See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/230659626

Empirical Bond-Strength--Bond-Length Curves for Oxides

Article  in  Acta crystallographica. Section A, Foundations of crystallography · May 1973

DOI: 10.1107/S0567739473000689

CITATIONS READS

1,380 2,261

2 authors, including:

Robert D. Shannon
University of Colorado Boulder
164 PUBLICATIONS   75,388 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

refractive indices and polarizabilities of minerals and synthetic compounds View project

All content following this page was uploaded by Robert D. Shannon on 25 February 2016.

The user has requested enhancement of the downloaded file.

266

Acta Cryst. (1973). A29, 266

Empirical Bond-Strength-Bond-Length Curves for Oxides
By I. D . BROWN AND R. D SHANNON*
Institute for Materials Research, McMaster University, Hamilton, Ontario, Canada

(Received 16 October 1972; accepted 4 January 1973)

Bond-strength-bond-length relationships for bonds between oxygen and H +, Li +, Be 2+, B a+, N a +,

Mg2+, AI3+, Sia+, ps+, $6+, K +, Ca2+, Sc3+, Ti4+, Vs+, Cr6+, Mn2+, Fea+, Fe2+, C02+, Cu2+, Zn2+,
Ga 3+, Ge 4+ and As s+ have been derived by requiring that the sums of the bond strengths around the
cations be equal to their valence in 417 crystals whose structures have been accurately determined. The
relationship is of the form s = (R/Ro)- N where s = bond strength, R = bond length and Ro and N are fitted
constants. It is further shown that all ions with an isoelectronic core can be fitted by a single pair of
parameters, Ro and N, that are independent of the ionic character of the bond and the coordination
number of the cation. The resulting bond strengths have the property that they are directly related to
the covalent character of the bond and that their sum around each atom is, on average, within about
5 % of its valence. The bond-strength-bond-length curves are particularly useful in accounting for
bonding in cases where the coordination is very distorted (e.g. Na +, Cu 2+ and Vs+). They can also be
used to predict the positions of hydrogen atoms, to analyze for different oxidation states and site oc-
cupancies, to calculate ionic radii and to provide an indication of the correctness of crystal structure
determinations.

Zachariasen (1954, 1963) and Zachariasen & Plettinger

Introduction
(1959) used the distances in a number of borates and
The concept of mean bond strength (~) was defined by uranates to prepare empirical tables of H +-O, Ba+-O,
Pauling (1929) as the valence (z) of a cation divided by and U6+-O ionic bond strengths vs. bond lengths and
its coordination (v). He enunciated the electrostatic these resulted in bond-strength sums which did not
valence principle by which the sums (p = ~ ) of mean differ by more than 0.1 valence unit (v.u.) from the
bond strengths around the cations and anions are ideal values. Clark, Appleman & Papike (1969) used
approximately equal to their valence. This principle an equation of the form R = a + bs + cs 2 where R = bond
has been tested frequently by mineralogists, particu- distance, s = bond strength, and a, b, c are variable pa-
larly in silicate structures, and is in general only approx- rameters for each ion pair to evaluate the strengths of
imately valid with the sum of the bond strengths around the bonds Na+-O, Mg2+-O, Ca2+-O, AI3+-O, Fe3+-O
the anions deviating by as much as 40% from the and Si4+-O in pyroxenes. Perloff (1970) assumed a
valence (Baur, 1970). linear relationship between bond length and bond
In the original formulation of the principle, the varia- strength for Mo6+-O distances in
tion of bond strength with bond length was determined Na3(CrMoO6024H6 ) . 8H20.
by the factor 1Iv. It is possible, however, to relate bond Donnay (1969) and Donnay & Allman (1970) have
strength (s) to bond length (R) using analytical ex- devised a scheme for constructing R - s curves using an
pressions ( R - s curves). Bystr6m & Wilhelmi (1951) equation of the form
used Pauling's (1947) logarithmic relationship for co-
valent bonds in the form: s= (2)
R - R~ = - 2k log n (1)
where s = strength* of a bond of length R;
between bond length, R, and bond number n,1" to eval- so=ideal strength of the bond of length /~, the
uate bond strengths in (NH4)2Cr207 and V2Os (RI= mean value of bond length in an individual poly-
length of a single bond and k is an arbitrary constant). hedron, and N = a constant which is different for each
Evans (1960) revised the value of R1 for V 5+ to analyze cation-anion pair and in some cases for different cation
valence sums in 10 vanadates containing V4+ and V 5+ sites. This expression is used for values of R </~. When
The agreement of calculated and observed valences was R >/~ they assume a linear relation between s and R,
for the most part excellent although the bond-strength
sum, Pv, around vanadium in V2Os was 5.47 and thus Rma x -- R
deviated considerably from the ideal value of 5.0. S = S0 -Rmax__ /~ (3)

* On leave of absence from Central Research Laboratory, * Although Donnay & Allman (1970) chose to use the
E. I. duPont de Nemours, Wilmington, Delaware. term bond valence, we prefer to continue Pauling's original
t Bond number is defined as the number of shared electron terminology. We have also changed the symbols in equation (2)
pairs per bond. to be consistent with those of Zachariasen (1931).
 I. D. B R O W N A N D R. D. S H A N N O N 267

Values of Rmax were determined for most metal-(O, OH) Procedure

bonds in the periodic table by extrapolating an empiri-
cal R - s curve derived from the ionic radii of Shannon The bond strength (s) can be calculated from the bond
& Prewitt (1969) to s = 0. They achieve excellent agree- length (R) using the computer program B O S T which
ment between the valence and the sums of the bond contains a variety of functions of the form s = f ( R ) ,
strengths but they are forced to construct unique curves [e.g. s = A((R - B)/c) -N ors = 10(A - R)/Ro + c]. We have
for each structure to be analyzed and in certain cases chosen to use the following expression
for even the same cation in different sites, e. g. Ca(I)
and Ca(2) in pumpellyite (Allmann & Donnay, 1971). S=So (5)
Baur (1970) uses the deviations from Pauling's elec-
trostatic valence principle to predict which bonds will because it gives a good fit and has the advantage of
be longer or shorter than the average in a particular simplicity. The parameter so (the bond strength asso-
crystal. He relates the bond length (Rtj) to the sum (p j) ciated with a bond of length R0) is assigned arbitrarily
of the mean bond strengths around the anion (j) using by the user and R0 and N are constants found by least-
a linear expression of the form square fitting as described below.
An initial set of parameters is chosen and used to
Rtj = at + bt pj~.
calculate the bond strength s~j for the bond between
He has determined the values of the empirical con- atom i and atom j using the observed bond length Rtj.
stants at and bt for fifteen M~-O atom pairs using the The sum (Pt) of these bond strengths around the atom
bond lengths observed in some 130 structures but he i is given by
requires different parameters t~or the same cation in p , = ~ s,j (6)
a different coordination. His approach works well pro- J

vided the range of Rtj values is not large, but for very
and in the ideal case this should be equal to the valence
distorted environments such as are sometimes found
around V 5+, his method does not work satisfactorily zt of atom i.
The initial set of parameters are then refined by
(Gopal, 1972).
least squares to minimize the function
We have derived a set of empirical bond-strength-
bond-length* curves based on the equation
Q= ~ wt(zt-pt) z (7)
t=l

S=So (4) where m is the number of atoms of type i used in the

calculation and wt is a weight set equal to 1/a2(pt), a(pt)
by fitting the constants, So, Ro and N, so that the sums of being the standard error in pl as calculated from a(Rtj),
the bond strengths around the cations are set as nearly the standard error in the experimental bond length.
as possible equal to the valence in a large variety of Refinement works best if the correlation between the
oxide structures. parameters Ro and N is kept small. This can be done by
While the equation we use [equation (4)] is the same choosing So so that Rtj/Ro~ 1; Ro then represents a
as that [equation (2)] used by Donnay & Allman (1970), typical bond length and so the corresponding typical
our approach differs from theirs in a number of signifi- bond strength.
cant ways. First, we use equation (4) over the whole The program calculates a number of measures of the
range of bond strengths, thereby avoiding the assign- agreement between the valence and the bond strength.
ment of Rmax (the maximum bonding distance) and the In addition to the difference, ( z t - P t ) , it calculates the
coordination number which they need in order to cal- relative difference (zt-p~)/zt and the difference normal-
culate/~. Secondly we use the same curve for all bonds ized to the standard error (z~-pt)/a(pt), the latter being
between two atomic species wherever they occur. the quantity which enters into the calculation of Q. The
Thirdly, we obtain the values of So, R0 and N in a overall agreement for the m atoms around which sums
different manner, using the information available in a have been made is conveniently measured by the r.m.s.
large number of crystal structures rather than tailoring relative deviation Dt = [(~(zt-pt)2/z~)/m] 1/2.
the values to the specific structures under consideration. Three different procedures were followed in deter-
mining the parameters Ro and N. For cations which
exist with several different coordination numbers bond-
* The concept of bond strength as defined by Pauling (1929) strength sums around all cations for which reasonably
is one derived from an ionic model of chemical bonding. In accurate bond lengths were available were calculated.
this paper we show that the concept can be used just as well in
situations where the bonding is primarily covalent (see below). The parameters were then refined by fitting the bond
The fact that the theory works in a given situation cannot be strength sums to the valence of the cation. In cases
taken as evidence that the bonding is ionic even though for where the cation normally has only one coordination
convenience we have continued to use terms such as 'cation' number this procedure is unsatisfactory since almost
and 'anion'. The term 'electrostatic bond strength' is used in
this paper in a formal sense only and it does not imply an ionic any value of N will give an equally good fit. In these
model. cases it was necessary to give much greater weight to
268 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

structures with atypical coordination numbers. For ex- Lazarev, 1964; Pant & Cruickshank, 1967; Brown &
ample the slopes of the R - s curves for 8i4+-0 and Gibbs, 1970; Shannon, 1971).
Mg2+-O were constrained to pass as close as possible Because this latter dependence is generally not yet
to the stishovite and SiP207 points and to the MgAI204 well defined, this effect has been ignored. The former
and MgV204 points respectively. Furthermore, since no dependence can be included by making an a priori
examples of either tetrahedral T i 4 + O 4 o r octahedral correction for the anion coordination number using
Ps+O 6 are known, we were forced to use data from the variation of oxygen radius observed by Shannon &
TiCI4, TiBr4, and PC15. In these cases Pauling radii for Prewitt (1969). Thus all distances can be 'corrected'
C1- and Br- corrected for CN in a manner similar to to those for two coordinate oxygen by subtraction of
that for 0 -2 (Shannon & Prewitt, 1969) were sub- the following amounts: 0.01A (3 coordination), 0-03 A
tracted from the average M - X distances. The resulting (4 coordination) and 0.05 A (6 coordination), etc. For
radii for 1VTi4+ and rips+ were added to the radius for comparison the bond strength parameters have been
1102- to obtain approximate lVTi4+-O and Vlps-o dis- refined using both the 'corrected' and the 'uncorrected'
tances. As in the cases of 8i4+-0 and Mg2+-O, the bond distances.
R-s curves were constrained to pass through the KS.lmS.n (1971) has shown that the average tetra-
lVTi4+-O and v~Ps+-O points. hedral M - O distances (/~) for atoms in groups IV to
The curves for H +, S6+ and C r 6+ w e r e refined in a VII of the periodic table can be expressed in terms of
different way which makes use of the fact that although the valence (equivalent to bond strength) by the simple
these atoms normally occur only with one coordination relation
number the bond lengths can vary considerably. In this /~=0"0113 n2 + R ' (8)
case the bond-strength sums around the oxygen atoms
as well as around the cations were used. This unfor- where n = 8 - z and R' is constant for a given row of the
tunately restricted the number of structures that could periodic table. This suggests that a similar relation
be uscd to those for which the bond strengths from the might also hold for the standard distances, R0. Rather
oxygen atoms to other cations in the structure could than use equation (8), we have found that we get a good
be calculated but on the other hand it gives a much fit with the function
stronger leverage on the value of N. The structures used
(R0) -~'~
in the refinements are listed in the Appendix. So= R1 (9)
Although the interatomic distances published in re-
cent crystal-structure determinations are frequently where Rx and N1 are constants for a given row. In most
quoted with standard errors of 0-005 A or less, it is well cases the value of NI obtained does not differ by more
known that they are subject to systematic errors larger than 1 from the individual N's of the atoms in the row.
than this. Principal among these is the error arising This allows a single pair of parameters to be used for
from thermal motion which is well understood but all the elements in the same row that have isoelectronic
difficult to apply because of the need to know how the ionic cores. Thus one curve of the form
atomic motions are correlated with each other. In cases
of molecular fragments which can be treated as rigid s0: (RRI 10,
(e. g. SO4z-) it is possible and customary to make these
corrections, but for more weakly bonded groups the can be used to calculate bond strengths around Li ÷,
correction becomes very uncertain. We have therefore Be 2+, and B3÷ and another to calculate bond strengths
only used distances uncorrected for thermal motion, around Na +, Mg 2+, Al 3+, SP ÷, p5+, and S6+ etc. The
conscious of the fact that these distances are system- fact that such a fit can be made implies that the length
atically too low (~0.01 to 0.02 A) and will have a true of the M - O bond of unit strength (R1) is the same (or
standard error much larger than the values quoted. A nearly the same) for all ions with the same electron core.
more critical evaluation of the effects of thermal motion To the extent to which this is true one can calculate
on the bond distances would, no doubt, improve the bond strengths for bonds not included in the table (e.g.
agreement between the bond-strength sums and the C-O, N-O, Cl-O; see, for example, Table 15).
valence, but such an evaluation is not possible at pres-
ent.
Results
The R - s curves are particularly useful in accounting
for the variation of bond lengths with cation coordina- Tables 1, 2 and 3 list respectively the refined param-
tion number, a large effect that has long been recog- eters obtained (1) from 'uncorrected' distances for each
nized (Goldschmidt, Barth, Lunde & Zachariasen, 1926; individual M - O pair (2) from 'uncorrected' distances
Pauling, 1927; Zachariasen, 1931). Recently, however, for ions having isoelectronic cores and (3) from dis-
the average distances have been found to depend in tances corrected for oxygen coordination for each in-
some measure on the anion coordination number (Jef- dividual M - O pair. Fig. 1 shows typical R - s curves
freys & Slaughter, 1963; Slaughter, 1966; Shannon & for Si 4+, Ge 4+ and Ti 4+. The points incidate the values
Prewitt, 1969; Brown & Gibbs, 1969) and the nature of of R that were used in the refinement and are, for con-
the cations surrounding the oxygen ions (Noll, 1963; venience, plotted along their respective curve. Fig. 2
 I. D . B R O W N AND R. D. S H A N N O N 269

s h o w s t h e curves d e r i v e d f r o m t h e u n i v e r s a l R - s pa- T a b l e 3. Bond-strength-bond-length parameters

r a m e t e r s a n d gives a g o o d i n d i c a t i o n o f t h e w a y in for M - O bonds (corrected for oxygen coordination)
w h i c h b o n d s t r e n g t h s fall o f f w i t h i n c r e a s i n g distance. used in the expression s = so(R/Ro)- N
Cation So (v.u.) Ro (A) N
T a b l e 1. Individual bond-strength-bond-length Li ÷ 0-25 1.925 3.4
parameters for M - O bonds in the expression Be2+ 0.5 1.611 3"8
S=so(R/Ro) -N B3+ 1.0 1-343 3"7
Na ÷ 0.166 2.421 5.7
Cation So (v.u.) Ro (A) N Mg z+ 0.333 2.076 5-2
H÷ 0.5 1.184 2.2 AI3÷ 0.5 1.888 4-6
Li + 0.25 1-954 3.9 Si4+ 1.0 1.605 4.0
Be z+ 0.5 1.639 4-3 ps+ 1-25 1.525 3-2
B a+ 1"0 1"375 3"9 K÷ 0-125 2.823 6.0*
Ca 2+ 0.25 2-437 5.5
Na ÷ 0.166 2"449 5"6
Mg 2÷ 0.333 2"098 5"0 Sc3÷ 0.5 2.090 6.0
A13+ 0.5 1.909 5.0 TP + 0.666 1-947 4"6
Si4+ 1.0 1-625 4"5 Vs ÷ 1.25 1.700 4.8
ps+ 1.25 1"534 3"2 Mn 2÷ 0.333 2.165 5.1
S6÷ 1-5 1-466 4.0 Fe 3+ 0.5 1.981 5-2
K÷ 0"125 2.833 5"0* Fe 2÷ 0-333 2-128 4-9
Ca z + 0.25 2.468 6.0 Co 2+ 0.333 2.087 5-0
Sc3+ 0.5 2"121 6.0 Cu 2÷ 0-333 2.068 5-4
Ti 4+ 0.666 1-952 4"0 Zn z + 0.5 1.946 5.7
V 5+ 1-25 1"714 5"1 Ga 3+ 0-75 1.822 5.2
Cr 6+ 1"5 1.648 4-9]" Ge 4÷ 1-0 1.735 5.3
Mn 2÷ 0.333 2.186 5.5 As 5+ 1.25 1.671 4-9
Fe 3+ 0.5 2.012 5.3 * Refines to N=4-7 but N = 6 . 0 gives as good agreement.
Fe 2÷ 0.333 2.155 5.5
Co z+ 0-333 2.118 5-0
Cu 2+ 0.333 2.084 5.3
Zn 2÷ 0.5 1.947 5.0 T h e relative m e r i t s o f t h e u n i v e r s a l a n d i n d i v i d u a l
Ga 3÷ 0.75 1-837 4.8 curves a n d t h e i n d i v i d u a l c u r v e s c o r r e c t e d f o r o x y g e n
Ge a + 1.0 1.750 5"4 c o o r d i n a t i o n c a n be j u d g e d f r o m t h e a g r e e m e n t in d ices
As 5+ 1.25 1.681 4" 1 (r.m.s. r e l a t i v e . d e v i a t i o n ) s u m m a r i z e d in T a b l e 4. C o m -
* Refines to N=4.3 but N = 5.0 gives as good agreement. p a r i s o n o f t h e i n d i v i d u a l vs. t h e u n i v e r s a l c u r v e s s h o w s
"1" Refines to N=6.1 but N = 4 . 9 gives better overall agree- b e t t e r a g r e e m e n t f o r t h e i n d i v i d u a l c u r v e s in 15 in-
ment. stances, a n d w o r s e a g r e e m e n t in o n l y 5 cases.]" T h e
m e a n r.m.s, d e v i a t i o n o f 4.0 % f o r t h e i n d i v i d u a l c u r v e s
T a b l e 2. Universal bond-strength-bond-length c a n be c o m p a r e d w i t h 5-4 % f o r t h e u n i v e r s a l curves.
parameters for M - O bonds in the expression Comparison of individual 'corrected' and 'uncorrected'
s = (R/RO- N1 c u r v e s s h o w s b e t t e r a g r e e m e n t for t h e ' u n c o r r e c t e d ' dis-
Number of t a n c e s in 10 cases a n d w o r s e a g r e e m e n t in a n o t h e r 12
electrons in cases; t h e m e a n r.m.s, d e v i a t i o n s b e i n g 4.6 a n d 4 . 4 %
Cations cation core R1 (A) NI respectively. Since c a l c u l a t i o n s i n v o l v i n g u n c o r r e c t e d
H÷ 0 0.86 2.17 d i s t a n c e s a r e s i m p l e r a n d since t h e i n d i v i d u a l c u r v e s
Li+Be2+B 3+ 2 1.378 4.065
Na+ Mg2+A13+Sp+PS+S 6+ 10 1.622 4.290 give a b e t t e r fit t h a n t h e u n i v e r s a l ones, t h e ' u n c o r r e c t -
K+Ca2+Sca+Ti4+VS+Cr6+ 18 1"799 4"483 e d ' i n d i v i d u a l c u r v e s w i t h p a r a m e t e r s listed in T a b l e 1
Mn2+Fe 3÷ 23 1.760 5"117 are to be p r e f e r r e d f o r m o s t a p p l i c a t i o n s .
Zn2+Ga3+Ge4+As 5+ 28 1.746 6.050
Discussion
The agreement between the valence and the bond-
s t r e n g t h s u m s is q u i t e sensitive to t h e v a l u e o f N ( + 0.2) 1. Electrostatic-valence principle
f o r small ions w h o s e c o o r d i n a t i o n n u m b e r s a r e well W e h a v e t r i e d to find a f u n c t i o n a l r e l a t i o n s h i p w h i c h
d e f i n e d e. g. Be 2+, B 3+, Si 4+, Z n 2+, G a 3+ a n d G e 4+. In u n i q u e l y relates b o n d s t r e n g t h to b o n d l e n g t h , r e g a r d -
o t h e r cases N m a y be v a r i e d w i t h i n a r a n g e o f + 2 less o f s t r u c t u r e t y p e , w i t h t h e c o n s t r a i n t t h a t t h e s u m
w i t h o u t m a k i n g an a p p r e c i a b l e difference to t h e agree- o f t h e b o n d s t r e n g t h s e q u a l s t h e v a l e n c e as p r o p o s e d
m e n t . * O n t h e o t h e r h a n d t h e values o f R0 are u s u a l l y
q u i t e well d e t e r m i n e d , v a r i a t i o n s o f 0.003 A p r o d u c i n g
I" The individual curves can always be made to give at least
a n o t i c e a b l e c h a n g e in t h e o b s e r v e d a g r e e m e n t . as good a fit as the universal ones since the individual param-
eters can always be chosen to be the same as the universal
* In certain cases when the refined value of N appears to be parameters. However, the method of fitting the parameters is
unreasonably large or small we have chosen a value different not one that necessarily minimizes the r.m.s, relative deviation.
from that obtained by the refinement procedure outlined above. One would expect the individual parameters to show system-
In all these cases, which are explicitly noted in Tables 1 and 3, atic trends across the periodic table and the universal curves
the agreement is equally good with both values. provide a reasonably successful attempt to describe these.

A C 29A - 4
270 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

by P a u l i n g (1929). This a t t e m p t to e v a l u a t e the electro- the b o n d lengths ( i n d i c a t e d by the errors given for t h e
static-valence principle follows the general lines a d o p t e d b o n d - s t r e n g t h sums) but s o m e h a v e to be a t t r i b u t e d to
by m o s t p r e v i o u s w o r k e r s but differs f r o m t h e m in t h a t real effects w h i c h we c a n n o t as yet explain. T h a t these
o u r data, e v a l u a t e d by least-squares t e c h n i q u e s , are effects are real is attested, for example, by the b o n d -
m o r e extensive. W e h a v e a n a l y z e d e n v i r o n m e n t s o f

20I
884 cations in 417 different structures a n d h a v e derived
R - s curves for M - O b o n d s for the m a j o r i t y o f a t o m s
in the first h a l f o f the p e r i o d i c table in a simple t w o -
p a r a m e t e r form. Table 4 s h o w s t h a t the m e a n r.m.s.
d e v i a t i o n o f the valence f r o m t h e c a l c u l a t e d b o n d - ~"
-i 1"5
s t r e n g t h s u m s a r o u n d c a t i o n s for i n d i v i d u a l curves is
i-
4.3 % ; the a g r e e m e n t is w o r s e for the ions Li +, N a +,
K +, Ca z+, Sc 3÷, a n d M n 2÷. Fig. 3, in w h i c h b o n d - t-
o
s t r e n g t h sums a r o u n d m o s t o f the c a t i o n s e v a l u a t e d in ~o 1.0
this study are p l o t t e d vs. t h e i r average b o n d length,
shows t h a t a l t h o u g h t h e m e a n d e v i a t i o n is a b o u t 5 %, O
133
i n d i v i d u a l v a r i a t i o n s o f 10-15 % are still observed.
A d e t a i l e d e x a m i n a t i o n o f the results for V 5+ shows 0"5
t h a t while o u r m o d e l a c c o u n t s for m o s t o f the gross fea- Ti
tures o f the structures significant d e v i a t i o n s still r e m a i n .
T a b l e 5 lists t h e b o n d sums a r o u n d the v a n a d i u m a t o m s i , , , I , , , , I

u s e d in o u r calculations. F o r c o m p a r i s o n the sums are 1"5 2.0 2.5

given using b o t h the i n d i v i d u a l l y fitted curve a n d the
curve for "cations" with the e i g h t e e n - e l e c t r o n core. Al- Bond Length (,~)
t h o u g h the average d e v i a t i o n f r o m 5.00 is o n l y 4.2 %, Fig. 1. Individual M - O bond-length-bond-strength curves for
it is a p p a r e n t t h a t t h e r e are i n d i v i d u a l d e v i a t i o n s o f M =Si, Ge and Ti. The circles indicate the values of the
individual bond lengths used in the calculations. They do
1 0 - 2 0 % in s o m e structures. S o m e o f these d e v i a t i o n s not represent the quality of fit between experiment and
can be a t t r i b u t e d to uncertainties in o u r k n o w l e d g e o f theory.

Table 4. Statistics concerning the derivation o f bond-strength-bond-length parameters

Method 1. Least squares fit R0 and N using cation sums only.
2. N chosen to give good fit for atypical coordination numbers. Ro refined by least squares.
3. Least-squares fit of R0 and N using cation and anion sums.
Number R.m.s. relative deviation (%)
Number of of bond Individual Universal Individual
structures strength uncorrected (Table 2) corrected
Cation used sums used Method (Table 1) (Table 3)
H+ 9 (3)* 38 (24)* 3 4.6 - -
Li + 22 29 1 6.4 10-5 5.3
Be2+ 11 13 1 3.4 4-0 3.7
Bs + 24 44 1 2"6 2"9 4"0
Na + 31 48 1 7-6 7"5 6-4
Mg z+ 27 39 2 5-2 4"9 5-1
A13+ 25 39 1 4"4 4"8 4"0
Si4+ 38 50 2 3"1 3"2 2"2
p5 + 29 45 2 2" 1 3"4 2"7
S6÷ 33 (3)* 41 (21)* 3 2"4 2"5 -
K÷ 29 34 1 8"5 10"9 9"0
Ca 2+ 30 40 1 6-3 8"4 6"6
Sc3+ 8 11 2 7"2 6"1 4"8
Ti 4+ 21 41 2 3-3 5-1 4"7
Vs + 31 46 1 4"2 4"1 3"9
C r 6+ 10 (5)* 22 (50)* 3 3"5 3"5 -
Mn z÷ 20 33 2 6-9 6"5 6"6
Fe 3+ 36 50 1 27 2"9 3"2
Fe 2÷ 15 19 1 5"4 - 4-8
Co z+ 18 29 2 6"6 - 3-7
Cu z + 25 40 1 5"8 - 6-5
Zn 2+ 29 48 1 4"8 4"9 3-9
Ga 3+ 10 14 1 3.2 6.6 2-8
Ge 4+ 17 28 1 3.7 5"4 4"4
As s+ 24 34 1 3.1 5"4 3"4
* Figures in parentheses indicate the number used in the refinement [marked with * in the Appendix]. The other figures indicate
the number of X-O structures and bonds used in calculating the indices quoted in the last 3 columns.
 I. D . B R O W N AND R. D. SHANNON 271

strength sums consistently found in a number of pyr- Table 5. Bond strength sums around V s +
oxene structures where Poe2) is typically 1.75 and
v = n u m b e r of bonds used in calculations given in subsequent
Poe3) lies between 2.15 and 2.30. Martin & Donnay columns.
(1972) attribute the low value of Poe2) to the presence of R =average bond length (A) around V 5+.
O H - ions. However, in the colourless pyroxenes LiVO3 Pt = bond strength sums calculated with individual R - s curve.
and NaVO3, small amounts of hydroxyl ion would be P I = ~ 1.25(R~/1-714) -s'l v.u. (Table 1)
t
accompanied by reduction of V 5÷ to V 4÷ which would P2 = bond-suength sums calculated with universal R - s curve.
probably result in darkly coloured compounds. It is P2= ~. (Rd1"799) -4"4s3 v.u. (Table 2)
t
likely that the second-nearest neighbours are respon- For references see the Appendix.
sible for some of these deviations but a detailed anal- Figures in parentheses are standard deviations in the last
figures quoted.
ysis of their role must await further study.
v R PI P2 Crystal
2. Bond-strength-bond-length curves and implications for 4 1"715 5"04 (7) 5"01 (6) Zn2V207
chemical bonding 4 1"706 5"12 (6) 5"08 (5) YVO4
4 1"721 4"90 (4) 4"89 (3) NdVO4
Bond-strength sums calculated using the 10 param- 4 1"729 4"84 (4) 4"83 (3) Mg3V208
eters of the five universal bond-strength curves give al- 4 1"722 4"91 (19) 4"90 (16) Co3V208
4 1"740 4"68 (18) 4"69 (16) Ni3V208
4 1"719 4"98 (7) 4"96 (6) Zn3V2Os
2.0 4 1"718 4"98 (16) 4"95 (14) Cd2V207
4 1"720 5"02 (5) 4"99 (5)
4 1"727 4"89 (5) 4"87 (4) FeV04
4 1"720 4"98 (5) 4"96 (4)
1.5
4 1"717 4"96 (1) 4"94 (1) Li3VO4
5 4 1"727 4"99 (2) 4'95 (2) LiVO3
4 1"730 4"92 (7) 4"89 (6)
4 1"731 4"84 (7) 4"83 (6) C02V207
== 4 1"707 5" 11 (4) 5"07 (3) Ca2VO4CI
I.o 4 1"691 5"36 (6) 5"28 (5) Na3VO4.12H20
O~
4 1"702 5"20 (8) 5"14 (7)
4 1"694 5"31 (10) 5"24 (9)
O
t~

0.5
\ \ 4
4
1"691
1"725
5"36 (9)
4"99 (23)
5"29 (8)
4"95 (20)
Ca3V2Os

NH4VO3
4 1"731 4"89 (7) 4"87 (6) KVO3
5 1"819 5"09 (7) 5"15 (6) CaV206
5 1"848 4"84 (22) 4"92 (19) "[
CaV206.2H20
0.0 5 1"808 5"44 (24) 5"45 (21) I"
I"0 I-5 2"O 2"5 3-0 5 1"952 5"12 (2) 5"12 (1)
5 1"883 4"81 (1) 4"89 (1) MgzV207
Bond Length (A)
5 1"830 4"99 (15) 5"06 (13) KVO3. H20
Fig. 2. Universal M - O bond-length-bond-strength curves for 6 2"009 5"29 (11) 5"36 (9) CsV3Os
isoelectronic series. The numbers associated with each curve 6 1.906 5.03 (2) 5.21 (1)
indicate the number of electrons on the 'cation' (core elec- 6 1.924 5.08 (3) 5-22 (3)
6 1.911 4.99 (2) 5.17 (2) Ca3Vl0Oz8.17H20
trons).
6 1.910 5"05 (2) 5"22 (2)
6 1.917 4"96 (5) 5.13 (4)
6 1.913 5"04 (5) 5-20 (5)
6 1.931 4.94 (6) 5.10 (5) KzZn2Vx002s. 16H20
6 1.914 4.85 (4) 5.05 (4)
6 1.932 4.94 (5) 5.10 (4)
6 1.947 5.25 (5) 5.35 (4) VPOs
!; 6 1.989 5.09 (7) 5.20 (6) V20s
6 1.917 5.27 (36) 5.38 (31)
6 1.908 5"64 (46) 5.67 (38) } VO(OCHD3
az 3 6 1.920 4.88 (12) 5-07 (10) CsV3Os
4.1 6 1-965 5.06 (3) 5"19 (3) COV206
t--
,o
6 1.969 5.13 (1) 5.25 (1) MgV206
6 1.944 4.96 (13) 5"12 (12) CdVzO6
2
"O
t-
O
t~ most as good a fit to the valence (5.4 % deviation as
x ~,...~÷ ..'.~," against 4.0 %) as those calculated with the 54 param-
eters of the 27 individual curves. The universal curves
thus lead to a considerable simplification in the concept
I I ! I
I.O 2.0 3"0 4"0 of bond strength for isoelectronic series of ions. It is
perhaps surprising that a single set of parameters can
A v e r a g e bond length (A) describe bonds ranging from almost completely ionic
Fig. 3. Experimental bond-length sums vs. average M - O bond to largely covalent, particularly as the bond-strength
length for most of the atoms used in this study. model of chemical bonding has always been regarded

A C 29A - 4*
272 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

as an ionic model. However, a prior knowledge of the other hand the long bond (R=2-787 /~,) has a bond
covalent character of a bond is not needed in order to strength of 0.103 v.u. and a covalence of 0.013 v.u. giv-
derive or apply the theory. On the other hand, there is ing 87 % ionic character.
a close correlation between the covalence of a bond as Hydrogen provides another interesting example of
calculated (Pauling, 1940, p. 72) from the electronega- the relationship between bond strength and covalence.
tivity difference between the terminal atoms and the Typically it forms two or more bonds between oxygen
mean bond strength between these atoms. Fig. 4 shows atoms in crystals. One of these is usually strong
the logarithm of the covalence plotted against the log- (s~0-85), the others are weak (s<0.15). The assign-
arithm of the mean cation-oxygen bond strength for ment of the electronegativity of hydrogen is based on a
'cations' with 18, 36 and 54 electrons. The relation- consideration of the strong hydrogen bond and leads to a
ship between the covalence (f~) and the bond strength covalent character of 67 % for a single O - H bond. In
(s) for M - O bonds can be described by the empirical crystals this covalence is shared very unequally between
equation the two or more bonds. Using a value of 1.8 to 2* for
f~=as M (11) the exponent M in equation ( l l ) the strong bond is
calculated to have about 58% and the weak bond
where a = 0 . 4 9 v.u. and M = 1.57. Similar curves can be 10% covalent character, according well with the view that
drawn for other isoelectronic series (Fig. 5) with the the weak O - H bond is primarily electrostatic. Both in
values of a and M given in Table 6. It follows from
equation (10) that the covalence is related also to bond * See Table 6. The conclusions drawn here are not very
sensitive to the exact value of M chosen.
length by
( R ) -MN1
f ~=a -~a (12)
2.0
with the values of a and MN1 given in Table 6. The
constancy of the values of MNt for the 'ions' with 2, 10
and 18 electrons is rather striking but may be fortu- Mo'C.~//f
itous.

Table 6. Parameters relating covalence to bond

strength and bond length in the equations
f'~=asM=a(R/R1) -MN1
1.0

0.t.
/
N u m b e r of electrons
in core a (v.u.) M MNI
0 0"67 1-8-2"0" 4*
2 0"60 1.73 7"04
10 0-54 1"64 7.04 "1. 0 ' 2
18 0"49 1"57 7.04 >
28 0"60 1"50 9.08 ~' 'Sc
36 0-49 1"57 ~" Py
46 0"67 1"43 ¢"
54 0"49 1"57
-~ 0.1
* Estimated from the trends in the value of M in subsequent >
O

/
series. ¢,.9

The relationships between the covalence and the 0"05

bond strengths of M - O pairs have been calculated on
the basis of mean values but it would not be unreason-
able to suppose that these relationships could be used
to calculate the covalence of an individual bond on the
basis of its known bond strength. This would be in ac- 0'02
/
cord with the notion that the covalent character of a
Rb
bond depends amongst other things on the oxidation
state and coordination number of its terminal atoms.
If these relationships can be applied to individual bonds I
they would allow a quantitative prediction of the differ- 0'01
ence in character between the very short and the very 0"1 0"2 0"5 I-0 2.0
long V-O bonds found in V2Os. The short bond ( R = Bond Strength (v.u.)
1.586 A) has a bond strength of 1.86 v.u. and a coval- Fig. 4. M - O covalence vs. bond strength for 'cations' with 18,
ence of 1.30 v.u. giving 30% ionic character; on the 36 and 54 electrons. The scales are logarithmic.
 I. D. B R O W N A N D R. D. S H A N N O N 273

strength and character the weak O - H bond is similar the 'hardness' of the core, larger values representing
to a typical N a - O bond, an observation that agrees harder cores. Both the properties of the core and the
well with the alkali metal-like behaviour of NH4+. * properties of the bond depend only on the numbers of
The fact that the relations between bond lengths, electrons in each, and since these are known from clas-
bond strengths and covalence are the same for all iso- sical valence theory, the prediction of bond lengths be-
electronic ions suggests a simple model of chemical comes remarkably simple.
bonding in which the electrons are divided into core This model is somewhat different from the ionic
electrons and valence electrons. The latter are all as- picture usually associated with a bond-strength for-
sociated with bonds, one electron pair per unit bond malism. In the ionic picture no distinction is made be-
strength, and are responsible for the attractive force tween core and valence electrons but the atoms are
in the bond. The core electrons give rise to the Born treated as negatively or positively charged spheres. In
repulsion between atoms. Equilibrium is achieved lo- our approach all atoms are treated as positively charged
cally when these two opposing forces are equal, the spheres (nucleus plus core electrons e.g. Li +, S6+ and
stronger the bonding force, the closer the distance of 0 2+ rather than O z-) held together by the negative
approach. valence electrons. In order to apply the theory it is not
The properties of the core are described by the fitted necessary to know whether the electrons are placed in
parameters given in Table 2 and illustrated in Fig. 2. the bond symmetrically (covalent bond) or asymme-
R1 describes the sum of the radii of the cation and trically (ionic bond), but the correlation between the
oxygen cores for a unit bond strength and N describes covalence and the bond strength suggest that the posi-
tion of the electrons with respect to the oxygen atom
* Khan & Baur (1972) have recently reviewed the similarities varies in a uniform way with bond strength, residing
and differences between NH4 + and alkali metals.
completely on the oxygen atom at zero strength (ionic
limit) and moving towards the centre of the bond for
20 bond strengths of about two (covalent limit).
Our model differs from the Valence Shell Electron
Pair Repulsion (VSEPR) theory of Gillespie & Ny-
holm (1957) in that we assume that the non-bonding
I'0 electron pair belongs not to the valence shell but to a

'~--'?
spherically symmetrical core. However, the VSEPR
theory is most useful in discussing angles subtended by
0.5 the ligands at the cation while our theory is more useful
in discussing distances.
18,,36, 3. Distorted coordination polyhedra
54
The stereochemistry of many elements can be discus-
sed quite effectively using a charged-hard-sphere model
>.
t-
o.z
281~"7/ in which the properties of an atom are described solely
by its ionic charge (valence) and its radius. Such atoms
tend to have rather symmetric coordination with bonds
(~ 0'I
> of equal length. The stereochemistry of such ions is not
o 4G- difficult to understand in terms of most theories of
chemical bonding. More recently the stereochemistry
of atoms which typically have rather distorted environ-
0"05 // ments (e.g. Cu 2÷, Sb 3÷ and 0 2-) have been discussed
,,//ii1"/;
in terms of the electron configuration in such theories
~,'"/"~O(H+) as the ligand field theory (Orgel, 1960) and the valence-
shell electron-pair repulsion theory (Gillespie & Ny-
holm, 1957). In still further cases (e.g. alkali metals and
V 5÷) distorted environments occur for which no really
predictive theory has yet been proposed. For atoms in
such distorted environments the concept of ionic radius
has proved difficult to apply and, because of the un-
0"01 certainty in assigning a coordination number, it has
0'I 0"2 0"5 I'0 2'0 also been difficult to use theories based on mean bond
strength.
Bond Strength (v.u.) The use of R - s curves resolves this problem by
Fig. 5. M-O covalence vs. bond-strength curves for the various avoiding the concept of coordination number alto-
isoelectronic series. The numbers associated with each curve
indicate the number of electrons in the 'cation' (core elec- gether. In cases where the coordination is regular, the
trons). The scales are logarithmic. bond strength for each bond can be calculated by divid-
274 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

ing the valency by the coordination n u m b e r (Pauling, Table 7. Bond lengths in selected sodium compounds
1929; Baur, 1970), and arguments based on average For reference see the Appendix
b o n d length or bond strength lead to the same con-
clusions. But the m e t h o d of calculating individual bond Ligand
number Na4P207 NaAI(SO4)2 Na2GeO3.6H20 NalO4
strengths can be applied just as easily in cases where the 2-453 2"36 2"55
1 2-271
coordination is very irregular. Two examples are given 2 2-314 2"453 2"37 2"55
in Tables 7 and 8 and are illustrated in Fig. 6 where the 3 2.326 2"453 2"43 2"55
average bond length and the bond-strength sums are 4 2.441 2-453 2"45 2"55
plotted as successively longer bonds are added to the 5 2.890 2"453 2-46 2-61
6 3.156 2"453 2-52 2-61
coordination spheres a r o u n d N a + and V s+. In the 7 3-294 3"68 2"61
bond-strength-sum graph, regular coordination appears 8 3-85 2-61
as a straight line and irregular coordination as a curved
line, but both level off at values of ~ s - - z. The inclusion
of very long bonds makes little difference to the bond-
Table 8. Bond lengths in selected vanadates
strength sums but alters the average bond length dra-
matically indicating that for these ions the concept of For reference see the Appendix
average bond length must be used with great caution. Ligand
In practice there will always be an arbitrary cut-off number ki3VO4 KVO~ V205 Ca3V1002,. 17(H20)
in the n u m b e r of bonds for which strengths are calcu- i 1.714 1-652 1.586 1.681
lated and this makes a small difference to the sums. At 2 1.717 1.661 1-782 1.681
3.0 A the bond strength for a N a - O bond is 0.05 v.u. 3 1.717 1.806 1"878 1.903
1.806 1.878 1.903
and inclusion of bonds in the range 3.0 to 4-0 A 45 !.720 2.023 2.135
corresponds to change in the value of R0 of about 0-01 6 2.787 2.135

2"0

2"8
1"9

2"7
o<
A

v
1"8
e-

r-
2"6 Li3V04 . (c)
m
ID NalO. _ _ ~ 1"7
t-

-0 O
2"5 (o)
V20~,/'
1"6

> Number of ligands

< 2"4

2'3 ,4 . . . . . . . . . . . .

0 Number of Iigands 8

1 1
j
. . . . . . . -
{b) I I!# <"
m 0 ~ : : ! ! I ~ I I
0 Number of Iigands 8

0 Numberof Iigands 8
Fig. 6. (a) Average Na-O bond length as successively longer bonds are included. (b) Bond-strength sums around Na as succes-
sively longer bonds are included. (c) Average V-O bond length as successively longer bonds are included. (d) Bond-strength
sums around V as successively longer bonds are included.
 I. D. B R O W N AND R. D. S H A N N O N 275

A. The parameters given in Table 1 were calculated Table 9. Variation o f mean M - O distance and
using a cut-off at around 3/~, the point at which N a - effective ionic radius in octahedral environments
cation distances begin to appear in the coordination as a function of distortion
sphere. This cut-off point also gives the best agreement The first equation given is that fitted to the observed values of
with the universal curves. R and A, the second (in parentheses) is that predicted from the
Baur (1970) is able to predict the variation of the R - s curves using equation (13).
The effective ionic radii (~) are calculated by subtracting the
bond lengths in distorted coordination by recognizing oxygen radius corrected for anion coordination from the mean
that where the sum of Pauling's average bond strengths bond length (Shannon & Prewitt, 1969).
6
a r o u n d the oxygen atom is greater than the valence, the A = ~ ~ [(Rt-/~)/1~]~ where/~ = average bond length in/~.
structure will compensate by weakening (lengthening) • /-1
Correlation Goodness
the b o n d to it. He obtains linear regressions between Atom max A Equation coefficient of fit
the a m o u n t of the lengthening and the excess bond Vs+ 0.0576 R = 1.887 + 2.62A 0.98 0.008
strength. Although his method gives a direct prediction (R = 1"855+ 5-8A)
o f the variation of individual bond lengths from the Cu z÷ 0.0316 /~ = 2-095 + 3"60A 0"89 0"016
(~ = 2"084 + 6"6A)
mean, it requires a knowledge of the coordination n u m - Mg2+ 0"0156 J~=2.094+ 8"31A 0"72 0"021
bers in the structure and a separate curve for a given (R = 2"098 + 6"3A)
cation for each coordination number. His linear re- =0"728 + 8-86A 0"77 0"018
gressions are equivalent to linear R - s curves and for Li ÷ 0"0148 R = 2"159 + 8-42A 0"31 0"030
the examples he discusses this approximation is reason- (R = 2"167 + 5"3A)
=0.784 + 9"02A 0"79 0"035
able since the average bond length is constant from Zn ~÷ 0.0071 J~= 2.099 + 7.7A 0.64 0.021
one site to another. On the other h a n d for strongly dis- (/~=2.111 + 6.3A)
torted coordination such as is found in octahedral V s+ = 0.736 + 8.20A 0.74 0"016
Co s+ 0.0046 /~=2.106+7-38A 0.42 0"019
and Cu 2÷ the average b o n d lengths show a significant (R = 2"118 + 6-4A)
variation. F r o m the non-linear shape o f our R - s curve =0.734+ 11-7A 0"70 0.016
we expect that such highly distorted coordination
would lead to a larger average b o n d length. The corre- The high correlation between the average b o n d length
lation between the average b o n d length (/~) and the and the distortion indicates that the V - O bond-
degree of distortion (A = mean-square relative deviation strength-bond-length curve, for example, could be fit-
f r o m the average) is shown in Fig. 7 for V 5+. Similar ted with a quadratic equation. Such an equation can be
data for Cu 2+, Mg z+ and Li ÷ are given by Shannon & obtained by expanding equation (5) by the binomial
Calvo (1973a, b). Linear regressions of average b o n d theorem to second order and this in turn leads to the
length on distortion yield the parameters given in linear relationship between average bond length and
Table 9 for V s+, C u 2 ÷ , Mg 2+, Li +, Zn 2+ and Co 2+. distortion:
For ions showing large distortion the correlation is / ~ = R , + R , ( N + 1) (13)
r e m a r k a b l y high. For only slightly distorted octahedra 2
other effects, such as anion coordination, become more
i m p o r t a n t and this is illustrated by the better fit that is where R, is the expected bond length for regular co-
obtained in these eases when the average bond length ordination. Table 9 list in parentheses the values for the
is replaced by the effective ionic radius which takes coefficients of equation (13) predicted from the R - s
anion coordination into account (Shannon & Prewitt, curves. The agreement with the fitted values is satis-
1969). factory where the distortion is small but for Li, Mg and
particularly Cu and V it is clear that the higher-order
terms in the expansion cannot be neglected. As these
depend on the nature of the distortion they will vary
from one cation site to another and cannot be calculated
A
2.0 for a general case.

4. Comparison with elastic properties

I
The assumption underlying the use of b o n d lengths
to measure bond strength is that the strength of a b o n d
can be measured by its ability to overcome the short-
range repulsive force that exists between any two atoms.
The parameter which measures this repulsion in an
I I I I I I I I I I ionic theory is the Born exponent (n).*
0'01 0~2 0"0;5 0'04 005

Distortion * Although the inverse power law is believed to be inferior

to the exponential form (Born & Mayer, 1932; Linnett, 1940;
Fig. 7. Average V-O bond length as a function of distortion Fumi & Tosi, 1964), Anderson (1970) has concluded that it is
(mean-square relative deviation of bond length from the satisfactory for analysing the pressure derivatives of the bulk
average) in VO~ octahedra. modulus of oxides.
276 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

Zachariasen (1931) has shown that in an ionic com- SiO2 and TiO2 anomalous and attribute the effect to
pound which exists in both the NaC1 and the CsCI bond bending and directional forces. The values of n
structure, determined from the elastic constant for BeO and for
CaO (Son & Bartels, 1972) differ by a factor of two
.-1 ~ nB
R= (14) from the values we have obtained. Fisher (1973) has
A zizze 2 suggested that the origin of this discrepancy may be in
and the derivation of n from the pressure derivative of the
Rcsct _ ,-1 1/i/nBcscl .AY_aC,
l (15) bulk moduli. Using the relationships derived by Ander-
RNaCI ~' /TBNaCt " Acsct son & Liebermann (1970), Fisher obtains values of n
for single-crystal MgO and CaO from the elastic mod-
where B = Born repulsion coefficient and A = Madelung ulus, C44, and the bulk modulus, K. These values agree
constant. Zachariasen assumed B to be proportional to well with the values of n determined from crystallo-
the coordination number of the cation and A not to graphic data in Table 10. It would be interesting to see
vary greatly from one structure to another. The latter if the discrepancy between the two values of n for BeO
assumption would appear to be fairly drastic but Tem- could be removed in a similar manner. In the light of
pleton (1953) has shown that in a variety of structures the rather drastic assumptions implicit in equation (16),
the reduced Madelung constant (~=average atomic the agreement between the Born exponent determined
Madelung constant) can be expressed as a function of from the elasticities and that determined from the crystal
average coordination number ( 0 in the form: structures is encouraging.
1
= 1.89- . ,
P Uses of bond-strength-bond-length curves
If ~ varies from 4 to 6, this results in only a 6 % varia- Previous workers have used bond-strength-bond-
tion in ~. Using equation (15) Zachariasen was able to length curves for a number of practical purposes related
rationalize the change in interatomic distance with co- to the evaluation of crystal-structure determinations
ordination number for a large variety of ions. Because (see Donnay, 1972). Our curves can achieve the same
Pauling's (1929) definition of bond strength merely in- results with considerably more generality and flexibility
volves z and v, it is apparent from equation (14) that as is illustrated in the examples given below.
z K
s= v = R,-- i . (16) 1. Testing for the correctness of a structure
Bond-strength curves provide a convenient method
The Born exponent can also be found from com- of testing the plausibility of a proposed crystal struc-
pressibility measurements. Anderson (1970) and An- ture, since the bond-strength sums around any atom
derson & Anderson (1970) have evaluated n using the should normally lie within 5 % of the valence. Larger
pressure derivative of the bulk modulus and their val- discrepancies in well refined structures might indicate
ues (calculated using the power-law repulsion term) are an incorrect structure. The effect is well illustrated by
compared in Table 10 with the values calculated from the structure of Zn3(BO3) 2which was originally refined by
the R - s curves. With the exceptions of BeO, CaO, Garcia-Blanco & Fayos (1968) in the space group Ic.
SiO2 and Ti02 there is satisfactory agreement. Ander- Baur & Tillmans (1970) subsequently refined it in the
son & Anderson have found the elastic behaviour of space group 12/c and although they obtained the same

Table 10. The Born exponent n evaluated from elastic constants and from bond-strength parameters
Born exponent
From individual From universal
From elastic bond-strength bond-strength
Compound constants curves§ curves§
BeO 9.5* 5-3 5-0
MgO 6"5,* 4"7t 6.0
A1203 5"0* 6"0
0c-SiOz 12'2" ]. 5"5 5"3
SiO2(stishovite) 14.0* J
MgA1204 5.6* 6"0
Mg2SiO4 6"5* 5"7
CaO 8.7,* ll'0,t 6-1+ 7.0
TiO2(rutile) 13"3* 5.0 5"5
Fe203 6"6* 6"3 6"1
ZnO 7.4* 6.0 7"0
* Anderson (1970) and Anderson & Anderson (1970)
"~"Son & Bartels (1972)
++ Fisher (1973)
§ n=N+l
 I. D. B R O W N AND R. D. S H A N N O N 277

agreement index ( R = 0 . 1 2 ) as Garcia-Blanco & Fayos (see for example Evans, 1960; Smith & Bailey, 1963;
they preferred their structure on chemical as well as Ribbe & Gibbs, 1969). Such methods depend on the
crystallographic grounds (Baur, 1970). The bond- construction of bond-length-site-occupancy curves for
strength calculations confirm Baur & Tillmans's pref- each pair of ions. The use of bond strength provides a
erence, giving a m e a n deviation of the bond-strength more general way of assigning site occupancy since such
sums from the valence of 13 % for the refinement in disordered atoms will normally have the same or very
Ic and only 3.4 % for that in I2/c (see Table 11). similar R - s curves, and the calculated bond-strength
sums will be equal to the average valence of the site.
Table 11. Bond-strength sums in two structures Two examples are given. In Table 12 the distribution
proposed for Zna(BOa)2 of A13+ and Si 4+ in sanidine, (Na, K)A1Si3Os, is pre-
dicted using the universal bond-strength curve, the
Space group Ic 12/c
Garcia-Blanco & Baur & bond-strength curve corrected for oxygen coordination,
z Fayos ( 1 9 6 8 ) Tillmans (1970) and ionic radii (Shannon & Prewitt, 1969). These are
Zn(1) 2 1.99 2-07 1.95 compared with the site occupancy determined from
Zn(2) 2 2.24 1.69 1"96 neutron diffraction and the agreement is quite satis-
Zn(3) 2 2.08 1.80 1.99 factory. The m e t h o d can be used with less precision to
B(1) 3 3.23 3-44 3"12
B(2) 3 2.76 3-67 3.01 calculate the ratio of K to N a on the alkali metal site.
O(1) 2 2-39 2-11 2.05 Table 13 compares the b o n d strengths in three ferro-
0(2) 2 2.08 1.66 1.95 spinels. The bond-strength sums at each site have been
0(3) 2 1.97 2.55 2.12 calculated with the R - s curves appropriate to each
0(4) 2 2.31 1.96 2-08
0(5) 2 1.82 1.95 1-91 species present and the weighted average bond strengths
0(6) 2 1.77 2.39 1.92 are shown for three cases-(a) n o r m a l spinel, (b) inverse
Mean deviation spinel and (c) r a n d o m spinel. In each case these are
from z 13 % 3"4% compared with the expected average valence. Table 13
clearly indicates that Fe2TiO4 is an inverse spinel and
2. Determining site occupancy in solid solutions FezZnO4 is n o r m a l in agreement with assignments
W h e n two atoms of different valence but belonging given by other workers. The result for Fe304 is not
to the same or neighbouring elements (e.g. A13+ quite as u n a m b i g u o u s but suggests that the Fe 2+ ions
and Si 4+ , Fe 2+ and Fe 3+) are present in a crystal, it is are disordered over all sites, a result at variance with
usually not possible to determine directly from X-ray accepted view that FeaO4 is an inverse spinel.
diffraction studies the ratio of each on any particular
site. The occupancy can be inferred from a study of the 3. Prediction of positions of hydrogen atoms
bond lengths between the site and its first neighbours The R - s curve for hydrogen has been calculated
since the lower valence ion gives rise to longer bonds using hydrogen positions determined by neutron dif-

Table 12. Cation distribution in sanidine

Structural information from Prewitt (1973)
Estimated errors given in parentheses.
Bond strength Universal Bond strength Ionic radii Neutron
sums bond-strength corrected for (Shannon & diffraction
Site (universal curves) sums O coordination Prewitt, 1 9 6 9 ) (Prewitt, 1973)
T1 3.72 72 (10) % Si 73 % Si 71%Si 68 % Si
T2 3.87 87 (10) % Si 82 % Si 92 % Si 82 % Si
K, Na 1.13 (as K) 62 (30) % K 88 % K
0.79 (as Na) 38 (30)% Na 12% Na

Table 13. Bond strengths in Fe2XO4 spinels

P(Fe a+) P(Fe ~+) P(X) P(normal) P(inverse) P(random)
X Site R(A) Table 4-2 Table 4-1 Table 4-2 Theor. Obs. Theor. Obs. Theor. Obs.
Ti A 2.00 2.01 2"49 4.00 2-49 2"00 2"01 2.67 2-17
B 2.03 2-78 3-49 2-00 2.78 3"00 3"16 2.67 3.02
Zn A 1.97 2.25 1.93 2.00 1.93 3"00 2.25 2.67 2.14
B 2.02 2-96 2.48 3.00 2-96 2"50 2.72 2.67 2-80
Fe A 1.887 2-80 2.59 2.00 2-59 3.00 2.80 2.67 2.73
B 2.056 2.69 2.76 3.00 2.69 2"50 2"72 2.67 2.71
References
Fe2TiO4 Ishikawa, Sato & Syono (1971).
Fe2ZnO4 Chalyi & Rozhenko (1956); Batanda, Szytula, Dimitrijevi6 & Todovovi6 (1969).
Fe3Oa Hamilton (1958)
278 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

fraction. In m a n y cases it is n o t possible to detect Table 15. Prediction of hydrogen atom positions
hydrogen atoms using X-ray diffraction and where it is in Na2CO3. H20 using universal bond-strength curves
possible their positions are not only less accurately
k n o w n but differ systematically from the positions Structural information from Dickens, Mauer & Brown (1970).
found by neutron diffraction. Consequently the bond Estimated bond
strength curves for H - O bonds quoted in Table 1 are Bond strength strengths for Estimated bond
only appropriate for bonds determined by neutron sums without hydrogen strength sums
H H(I) H(2) with H
diffraction.
Na( 1) 1.02 1-02
Often it is of interest to know where the hydrogen Na(2) 1.11 1.11
atoms are to be f o u n d in the crystal in those cases where C 3.98 3.98
their presence is not detectable by X-ray diffraction. O(1) 1.88 0.16 2.04
Frequently the presence o f short O - O distances is in- 0(2) 1.98 1.98
0(3) 1.80 0.24 2.04
dicative of a hydrogen b o n d and the hydrogen atoms 0(4) 0-45 0.84 0.76 2.05
can be assigned qualitatively without much ambiguity. 1.00 1.00
D o n n a y & Allman (1970) have shown that the calcu- Hydrogen bond
lation o f bond-strength sums provides additional in-
Bond Bond Predicted Observed
f o r m a t i o n about the location o f hydrogen atoms. Our strength length (A,) O-H" • •O O-H" • •O
b o n d strengths can be used in a similar way but have the
O(4)-H(1) 0-84 0.93 3"0 2.907 (2)
additional advantage of ease in calculation. Table 14 H(I)-O(I) 0.16 2.1
shows the calculations of b o n d strengths for O(4)-H(2) 0.76 0.96 2.7 2-684 (2)
M g S O a . 4 H 2 0 both with and without the inclusion of H(2)-O(3) 0.24 1.7
the hydrogen atoms. Table 15 shows similar calcula-
tions for Na2COa. H 2 0 where the hydrogen positions
are not known. Estimates o f O - H distances based on Table 16. Effective ionic radii oJ" ions in unusual
predicted b o n d strengths give values for the O - H - • • O coordinations
distances that are in striking agreement with those ob-
served. A somewhat different prediction for the posi- Ion CN Radius
tions of the hydrogen atoms based on different con- Li + V 0.69
B3+ VI 0-27
siderations has recently been made by Baur (1972). ps+ VI 0.38
S6+ VI 0.29
4. Calculation of effective ionic radii V5+ V 0"43
The R - s functions given in Table 3 allow one to ob- Vs+ VI 0"50
Mn 2+ IV 0.65
tain more statistically valid effective ionic radii than Mn 2+ V 0.74
were previously possible and in addition allow extrap- Fe 2+ V 0.70
olation to obtain radii for which no examples have yet Fe 2+ VIII 0.91
been found. Some examples of previously undetermined Fe 3+ V 0.56
Ge 4+ V 0.46
radii and o f radii differing significantly from values in As s + V 0"40
S h a n n o n & Prewitt (1969) are listed in Table 16. As5+ VI 0"46

Table 14. Bond strengths in MgSO44H20 calculated using universal curves

Structural information from Baur (1964)
Bond strength Bond
sums Bond strengths for hydrogen atoms strength
without H H(I A) H(1 B) H(2A) H(2B) H(3A) H(3B) H(4A) H(4B) sums with H
Mg 2-14 2.14
S 6.08 6-08
o(1) 1.84 0.16 2"00
0(2) 1.89 0.07 2"05
0.09

0(3) 1"50 0"15 0.17 0.19 2"00

0(4) 1"55 0.14 0-12 0.15 1 "97
O(W1) 0"37 0.82 0"86 2"05
O(W2) 0"35 0.82 0.90 2"07
O(W3) 0"36 0.86 0.78 1 "99
O(W4) 0"36 0.84 0.79 1 "99
Bond strength
sums for H 0.97 1.03 0.96 1"06 0.98 0.94 0"99 0.98
O-H. • .O
distance (A) 2.884 2"754 2.835 3.282 2.860 2.833 2.831 2.734
3"042
 .I.D. BROWN AND R. D. S H A N N O N 279

5. Testing for positional disorder or anomalous thermal o f 1.56 A normally found in these crystals. The large
motion calculated bond-strength sums in this case are thus in-
As mentioned above, thermal motion tends to de- dicative o f either a positional disorder or an anomalous
crease the apparent bond length as determined by dif- thermal effect.
fraction methods. In the present discussion no allow-
ance has been made for this effect on the assumption Table 17. Bond-strength sums around the bridging
that the extent o f the shortening will be approximately oxygen atoms in pyrophosphate groups
the same for all bonds. If, however, the thermal motion
is anomalous large (or small) the apparent bond lengths Bond-strength
sum P-O-P
will be shortened (or lengthened) from the values used (Table 4-2) (o) Reference~
in this study, giving rise to anomalously high (or low) Cd2P207 1.99 (5) 132 69 CJCHA 47 3409
bond strengths. The effect is well illustrated by the Na4PzO7 2"10 (1) 128 72 CJCHA 50 2519
bond-strength sums calculated around the bridging ~-Zn~P207 2"10 (2) 139 70 JSSCB 1 120
~-Sr2P207 2"12 (5) 131 68 ACSAA 22 1419
oxygen atoms in pyrophosphates (Table 17). All the 66 ACCRA 21 942
fl-Ca2P20~ 2.12 (4) 131
sums are somewhat high but they tend to increase as the ct-Ca2P~07 2.14 (3) 130 68 INOCA 7 1345
P - O - P angle increases reaching values between 2.3 and o~-Mg2P207 2.18 (4) 144 67 ACCRA 23 289
2.4 in those structures where the P - O - P angle is con- fl-Ca2P207 2-22 (4) 138 66 ACCRA 21 942
~.-Zn2P207 2-23 (3) 148 70 JSSCB 1 120
strained by symmetry to be 180 °. In these latter cases ~-Cu2P207 2"26 (3) 157 67 ACCRA 22 665
the thermal parameters o f the bridging oxygen atom SiP207-III 2"27 (4) 139 70 ACBCA 26 233
are large, indicating that the atom is normally dis- ,8-Zn2P207 2-31 (2) 180 65 CJCHA 43 1147
ordered away from the centrosymmetric position. The fl-Mg~P~O~ 2"38 (2) 180 65 CJCHA 43 1139
instantaneous P - O - P angle is thus less than 180 ° and i The references are given in the same form as in the Ap-
the corresponding P - O distances longer than the values pendix.

Table 18. Structures used in the evaluation of bond-strength-bond-lengths parameters

H(lb~ 87 ACCRA 22 37 NA2 SZ 03
69 ACBCA 25 313 BE ( H 2 . 0 ) ' S . O ' 68 ACBCA 2' 13 ALPHA-qAZ S I 2 05
~66 ACCRA 20 21..D B ( 0 . 0 ) 3 88 ACBCA 24 1u77 BETA-NA2 SIZ 05
69 ACBCA 25 180h CD [N.33) 2 {02.0) h 71 ACBCA 27 2 1 Z h HA3 P.3~ [HZ.O}%Z
66 ACCRA ZO 642 I H (N.O3)(* (HZ.O) 9 CJCHA REF2 NA, P2.07
~87 ACCRA 22 182 NA AL { S . 0 ~ ) 2 ( H 2 . 0 ) 1 2 87 ACCRA 22 43 NA4 P2°07 [ H 2 . 0 } 1 0
86 ACCRA 21 383 CA AL ( S . 0 , ) 2 H2.0 85 ACCRA 18 ZO NA3 P3 09
66 ACCRA 21 705 NI S.O, ( 0 2 . 0 } 6 69 ACBCA 25 1077 NA5 P3 08 (HZ.O)I~
89 ACBCA 25 o21 SHM (BR. 03)3 (H2.0)9 70 ACBCA 26 158(* NAg H oeO, [H2.0) 7
66 ACCRA 20 765 c 1.05 71 ACBCA 27 29% NA2 H% HP2 06 ( H 2 . 0 ) 8
LI(I~') 67 ZEKGA 125 298 CA2 NA SZ3 09
88 JCPSA ~ 3348 GAMMA-LI B 02 85 ZEKGA 122 175 CA NA2 ($o0(*) 2
71 ACBCA 2 9U(* L I 3 B°33 71 ACBCA 27 212(* HA3 V. OW [H2.0}12
85 ACCRA 19 =.81 L I C8 47 07 70 ACSAA 24 3627 HA% CR] 08 O.H
67 ACCRA 22 906 L I NA 5 ° 0 , 70 INOCA 9 Z228 NA3 (CR MO 06 02,, Hb) ( H / . U ] 8
69 ACCRA 19 39~ GAMMA L I AL 02 71 MOCHA 102 96~, NA2 GE 03
68 ZEKGA 126 L I AL 3IZ Ob 67 ACCRA 22 507 NA2 GE 03 ( H 2 . 0 ] b
61 ACCRA t~ 399 L I AL S I , 010 ?0 ACSAA 2~ 1287 NA4 SNZ GEt* 012 ( O . H ) ,
67 INOCA 119 L OH-LI] P . O ' 87 ACSAA 71 1."81 NAB SN~ G E t 0 030 (O,H),
70 ACSAA 24 2803 11 H% ° . 0 3 71 ACBCA 27 212(* NA3 AS.O(* ( H Z . O ) 1 2
02 ACCRA 15 l d B J T KZ ~)5009 H2mO UNPU1 REF3 NAt* AS2.07
88 CIWYA 88 Z90 LT FE ° ° 0 4 70 ACBCA 26 157(* NA2 H A S . O , ( H 2 . 0 ) 7
60 ACCRA 13 325 t i I MN P . O , 70 ACBCA 28 %584 NA2 H AS.O(* ( H 2 . 0 ) 7
65 ACCRA 18 71T S.O:+ H2. 8 70 ACSCA 28 1782 NA I.OC*
61, ACCRA 17 65t* L I N2..45 S° ' MG(2*)
UNPUI REF% L I V.03 57 AMMIA 42 39 MG 0
UNPUI REF% LT3 V.O~ bb NJMHA 1966 1(.2 MG 84.37
69 MSAS% 92 3% L I FE ~IZ 06 71 AMHIA 58 1553 MG 88 )Z (O°H)6 (H2.0,2
71 ZAACA 37 157 L I 2 CU OZ 85 MIASA 1985 198 HG AL ,.Ok
85 ACCRA 18 ,81 L I GA 32 59 AMHIA 44 679 CA HG {C. 03)2
70 ACSAA 24 3711 L I MO.02 AS.O(* 69 SPHCA 13 599 MG AL20U
88 JPCSA 27 997 t ~ NB 33 65 AMMIA 50 2U23 MG3 AL2 SI3 012
67 JPCSA Z8 l o 8 5 NB 03 88 ACBCA 24 1518 HG AL3 5I 9 09
87 JPCSA 28 1693 L I MB 03 69 ZEKGA 129 65 MG SI 03
71 ACSAA 25 3337 L% NB3 08 65 THPMA 10 71 CA HG 31.0(*
67 JPCSA 28 1893 L I TA , 3 68 AMMIA 53 80 MGI.8 FEO.Z SI.O~
67 JPCSA 28 1285 L I TA 33 67 ACCRA 23 289 ALPHA-4G2 P 2 , 0 7
70 ZAACA 37~ 206 L I EU30~ 68 ACSAA 22 1,88 HG3 P% 08
BE(2~'} 87 ACCRA 23 (.18 MG H P.O(* (HZ,O)a
84 JESOA 111 78 BE 0 70 ACBCA 28 1429 MG N.HC~ P . O , ( H 2 . 0 ) 8
63 ACCRA 16 1 1 4 ~ BE% 6.33 O.H 70 8SCFA 12 h243 HG S*O+ H2oO
63 AMHIA 48 80(* BE AL2 OW 6~ ACCRA 17 883 MG S,OW (HZ.O]4
62 ZEKGA 117 18 AL BE ~ I . 0 4 O°H 8~ ACCRA 17 235 MG S.O(* (H2°O)b
69 ACBCA 25 791 Y% SI ~EZ 07 84 ACCRA 17 1361 MG S.O'* ( H Z . O ] I
71 SPHCA 15 999 FE3 BE S I 3 09 (U.H~F)Z 82 ZEKGA 117 34, MG ( N . ~ ' ) 2 (S.04J2 (H2.0)8
62 ZEKGA 117 16 [Hb~FE) BE P.O c• O.H 5~ ACCRA 7 246 HGb MN 08
89 ACBCA 25 310 BE S.O(* ( H 2 . O ) , 69 ZAACA 369 308 HG V% 34
69 ACBCA 25 1 8 , ? S F BE3 OW UNPU$ REF3 HG2 V% O?
67 ACBCA 2(* 872 LAZ BE2 35 71 GJCHA 49 1829 HG3 V% 08
87 ACCRA 22 55~ Y% BE.O, 70 JSSCB 2 812 MG28 G'IO 0 , 8
B(3<') 70 CJCHA 48 890 HG2 AS% 07
70 ACBCA 26 906 82 03 I 71 ACBCA 27 815 MG3 TE 36
88 ACBCA 2(* 889 82 03 II 69 SPHCA 13 933 MG W.Ow,
83 ACCRA 10 ~89 BETA-H B°02 AL(3t]
83 ACCRA 16 380 GAMMA-H B.OZ 82 ZEKGA 117 235 AL2 03
5~ ACCRA 7 305 H3 B.O$ 85 ACCRA 19 396 GAMMA L I AL 02
66 JCPSA 3348 GAMMA-LI B.OZ 68 ZEKGA 128 46 L I AL ;12 06
71 ACBCA 27 90~ L I 3 8 33 83 AMMIA 48 804 BE AL2 04
63 ACCRA Ib 5~ NA BoO- ~ 65 MIASA 1 9 ~ 19~ AL HG 3.0~
63 ACCRA 18 1233 NA 8 ( 3 , H ) ' (H2.0)2 68 NJHMA 1988 80 CA AL 3°0(*
65 AMMIA 50 1827 HA 8 SI3 08 68 ACBCA 2~ 1518 MG AL3 SI B 09
70 ACBCA 28 1189 K B 02 86 AMMIA 51 956 NA AL $12 06
63 ACCRA 18 37o K 85 08 (H2,.O)~ 87 ACCRA 22 182 NA AL [S. 0 , ) 2 (H~:.O]IZ
6~ ACCRA 16 975 MK~ B" 37 (H2,~0)4 70 Z E K G ' 131 31(* NA CA AL SI2 07
MIASA 1985 198 AL 9.04 85 AHMIA 50 2023 MG3 ALZ SI3 012
88 ACBCA Z(* 1518 MG AL3 $1 B 09 81 ZEKGA 115 269 '~ANDALJSITE* AL2 SZ 05
63 ACCRA 16 390 CA B2 )W [ I ) 83 ZEKGA 118 337 ~KYANITE" ALZ SI O~
89 ACBCA 25 95~ CA B2 )W ( I I I ) 83 ZEKGA 118 127 '*SILIMANITE" ALZ SI 05
69 ACBCA 25 965 CA B2 Ok ( I V ) 86 ZEKGA 123 81 CA3 AL-~ S I 3 012
89 NJMMA 1969 Z(*d CA3 B2 DO 66 ZEKGA 123 181 AL P,O*
70 ZEKGA 132 2,~ CA B3 05 (O.H} 68 AHHIA 53 1096 AL2 P.34 [ O . H ) 3
68 NJMHA 1988 80 CA AL 9 . 0 4 85 ACCRA ~.9 971 Y ~ AL5 012
67 ZEKGA %25 286 CA B S I . O 4 0 65 ACCRA 19 97% GD3 AL5 012
69 ACBCA 25 1811 8A [B (O°H] 'Hz H2.O 65 ACCRA 19 971 YB3 AL3 012
71 ACBCA 27 677 CU BE 34 65 ACCRA 19 971 LU3 AL5 012
NA(it) 71 AMNIA 56 ]91 * ILMAN)INE GARNET~'
87 ACCRA 22 906 NA L I 5.04 71 AMMIA 58 791 *MN-GROSSULAR GARNET*
83 ACCRA 18 594 NAB OZ 71 AHMIA 58 791 "$PESSARTINE GARNET~
63 ACCRA 16 1233 NAB (3.H)4 (H2.0)2 Sl(k~)
65 AMMIA 50 1827 NA B S I 3 08 85 ZEKGA 121 369 ~CRISTOBALITE ~ SZ 02
62 ACCRA 15 77 NA H C,03 65 ACCRA 18 710 ~'QUARTZ*' SI 02
bb AMHIA 51 956 NA AL SI2 08 71 ACBCA 27 2133 *STISH3VIT."* SI 02
87 ACCRA 22 182 NA AL (S, 04)2 (H2.0)12 68 ZEKGA 128 46 L I AL $I2 06
280 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

We wish to thank Dr K. K. Wu for calculation of the ANDERSON, O. L. (1970). Y. Geophys. Res. 75, 2719-2740.
parameters of many of the R - s curves, Mr P. S. Gu- ANDERSON, O. L. 8c LIEBERMANN,R. C. (1970). Earth Planet.
merman for assistance with the figures, Mrs Ruth Interiors, 3, 61.
Shannon for preparing the bibliography and Drs D. B. BALANDA, M., SZYLULA,A., DIMITRIJEVIC', Z. & TODOVO-

Rogers, W. H. Baur, G. Donnay and D. T. Cromer for Vld', J. (1969). Phys. Stat. Sol. 32, K91-K93.
BAUR, W. H. (1964). Acta Crvst. 17, 863-869.
critical comments. The research was supported by the BAUR, W. H. (1970). Trans. Amer. Co,st. Assoc. 6, 129-
National Research Council of Canada. 155.
BAUR, W. H. (1972). Acta Cryst. B28, 1456-1465.
APPENDIX BAUR, W. H. & T~LLMANS, E. (1970). Z. Kristallogr. 131,
213-221.
Structures and their references used in the evaluation BORN, M. & MAYER, J. E. (1932). Z. Phys. 75, 1-18.
of the bond-strength-bond-length parameters given in BROWN, G. E. & GIBBS, G. V. (1969). Amer. Min. 54, 1528-
Tables 1, 2 and 3 are given in Table 18. The reference 1539.
is given in the following form: year, journal (a list of BROWN, G. E. & GIBBS, G. V. (1970). Amer. Min. 55, 1587-
1607.
codes used is given at the end), volume and page num- BYSTROM, A. & WILHELMI, K. A. (1951). Acta Chem. Scand.
ber. For papers marked with an asterisk, see Table 4. 5, 1003-1010.
CHALY[, V. P. & ROZHENKO,S. P. (1956). Dokl. Akad. Nauk
References SSSR, 108, 875-877.
ALLMAN,R. & DONNAY,G. (1971). Acta Cryst. B27, 1871- CLARK, J. R., APPLEMAN,D. & PAPIKE, J. (1969). Miner.
1875. Soc. Amer. Spec. Paper, 2,31-50.
ANDERSON, D. L. & ANDERSON, O. L. (1970). J. Geophys. DICKENS, B., MAUER, P. A. & BROWN, W. E. (1970). J. Res.
Res. 75, 3494-3500. Natl. Bur. Stand. A 74, 319-324.

Table 18 (cont.)
TI(~*) 62 JINCA 2W 23 MN CR2 Ok
55 JACSA 77 Wz08 ~ANAST~SE ~ TI 02 69 ZEKGA MN3 FEZ GE3 012
71 JCPSA 55 320o ~RUTIL-'~ TI 02 1:>9 427
69 ZE~GA 129 222 69 ZEKGA 130 139 HN GE 03
BA TI ~I3 09 68 ZAACA 358 125 MN SE 3~
b9 ZEKGA 130 438 BA2 TI 0 SIZ 07 70 PEPIt 3 161 BETA M~2 GE,OW
6w ACCRA 17 2WO CO T I 33 b8 AMHIA 53 73~ MNT (O.H) 8 AS2 08
68 ACBCA Z~ 132~ Y2 TI 35 ?0 AMMIA 55 2023 MN9 (O.H) 9 ( H 2 . 0 ) 2 A $ . 0 3 AS2 08
70 ZEKGA 131 Z78 YZ T I 2 O? 62 ZEKGA 117 331 MN HO 3W
66 CHDAA 262 96~ LA2 TI 05 65 JCPSA ~,3 2533 HN HO Ok
05 CJCHA W3 Z~IZ ER2 TIZ O1 FE(3#)
69 CJCHA k7 971 G02 T I 2 07 66 AMMIA 51 123 ALPHA FE2 03
69 CJCHA W7 971 LU2 TIZ 07 69 NSAPA Z 31 L I FE S12 06
69 CJCHA W7 971 SM2 TI~ O? 67 ACCRA 22 ?66 NA3 FEB 09
O0 JCPSA 32 1515 BA T I ~ 09 69 HSAPA Z~ 31 NA FE ~IZ Ob
ACBCA 25 1~ BA T I 5 01 152q NA FE3 83 AL6 Sit> 030 F
CRSCl 1 BA T I 6 013i 6? AH
MA
IA
CBCA 5() 191~ K ;[ (3,0W|2
70 ACBCA 2~ 16~5 BA6 T117 OWO 65 AMMIA 50 1929 {S.0~)2 H2,0
ACCRA 21 9ZW TH T I Z 06 71 ,qNI' 5b 1917 K FE (~.04)2 (HI.O)4
55 BCSJA 29 95 T I CL, 68 ZEKGA 127 261 F 2 (S. OW)2 0 ( H 2 t 0 ) 7
TFSOA 37 393 T I LR~ 70 A~BCA 26 1~,69 C~2 FEZ D5
V(5+) 72 SSGOA 10 5 CA3 FEZ GE3 012
bl ZEKGA 115 11d ~'~ 05 71 HRBUA b 725 BA CA FEW 08
UNPU1 REF1 L I V.03 69 ZEKGA 129 ~27 MN3 FE.~ GE3 012
UNPU1 REF1 L13 V.OW 71 JSSCB k 1 FE V O*
CZYPA W'8 (0 ~ : H 3 ) 3 65 JCPSA W2 3957 G A ( Z - X ) F E t X ) 03
66 INOCA ~ 2131
120 EU3 FEZ. GAB 012
60 ZEKGA 11W 257 N : H & ' ( ~ . 0 3 H3)3 ~8 ACBcAJCPS' 48 1094
26 2008 ND FE 03
71 ACBCA Z? 212~ NA3 V.OW (H2.0)12 70 ACBCA 26 2008 PR FE 33
CJCHA REFo MG V2 Ob 70 ACBCA 26 2008 SM FE )3
UNPUI REF3 MG2 V2 O? 70 ACBCA 2b ZU08
71 CJCHA ~9 1629 HG3 V2 08 70 ACBCA 26 2008 FE 03
JSSCB REF~ V P 05 70 ACBCA 26 2008 TB FE 03
ZEKGA 114 257 K V,03 70 ACBCA 26 2008 DY FE '33
ZEKGA 11W 257 V,O3 H2°O 70 ACBCA 26 2008 HO FE O3
BSCFA 3 736 K V*03 H2,,O 70 ACBCA 26 2008 ER FE 03
66 INOCA K2 ZN2 VlO 028 ( H 2 t O ) l b 70 ACBCA 26 2008 TM FE 03
?0 INOCA % 2259
967 CA2 V,OW CL 70 ACBCA 26 Zd08 YB FE 03
70 CHOCA 27~ 952 CA V2 36 70 ACBCA 2b 2008 LU FE O3
CAMIA ~8 CA V2 06 ( H 2 , 0 ) 2 65 ACCRA 19 9;'1 Y3 FEB 012
ZEKGA REFW CA3 V2 08 65 ACCRA 19 971 SM3 FEB 012
66 ACCRA 21 397 CA3 V l J 028 ( H 2 , 0 ) 1 7 61 ACCRA 1W 1051 GO3 FEB 012
JSSCB ~ 29 FE V.O~ 70 SSCOA 8 1745 TB3 FEB 01Z
UNPU1 REF3 CO V2 06 65 ACCRA 19 971 D¥3 FE5 012
UNPU1 REF3 C02 V2 07 70 SSCOA 8 1 ?w5 H03 FEB 012
~NPU1 REF~ C03 VZ 08 65 ACCRA 19 971 YB3 FEB 012
UNPU1 REF3 N I 3 V2 08 65 ACCRA 19 971 LU3 FEB 012
UNPU1 REF3 ZN2 V2.07 FE(24}
CJCHA W? 3056 ALPHA-ZN3 V2 08 67 JCPSA W7 (*559 FE 0
54 ACCRA 801 CD VZ 06 66 NJHMA 1966 362 HW FEZ AL 5 I ~, 02~,
69 BSCFA 3 736 CO V2 06 (:,9 NJMMA 19b9 43d FE ALZ P2 08 (O.H)Z
67 CJCHA ~5 2291 CD2 V2. O? 71 SPHCA 15 999 FE3 BE S I 3 09 (F~O.H)2
68 ACBCA 2~ 292 Y V, OW 67 CIWYA 67 285 CLINO-F~ SI 03
70 UNPU1 REF7 NO V°04 67 CIWYA 67 285 ORTHO-FE 51 03
66 INOCA 5 1806 CS V3 08 b5 ACCRA 18 787 FEZ SI Ok
58 AMMIA 53 807 F E2 SI Ok
ACBCA 26 2ZZ CR 03 69 $CIEA 166 1399 (NAtK}2 FEW SI12 030 H2.0
70 ACSAA 24 211 (PH3 SI)Z CR~O~ 69 CIWYA 68 290 FE LI P.OW
?0 AHMIA 55 1103 ZN PB19 F2 (CR, OW)6 (SI,OW)2 70 BSCFA 93 190 ALPHA-FE S.O&
~68 CJCHA ~6 933 KZ CRZ. OT 62 ACCRA 15 815 FE S°Ow (H2°O)W
69 BUFCA 92 ~bh CO CR, O~ 67 ACCRA 22 775 FE (N.HW) 2 (S.OW)2 (H2.0)6
69 8UFCA 92 2bh NI CR.04 71 JUPSA 31 ~52 FEZ TI OW
69 JCSIA 1969 1857 (N,H~.)_~ CR, O~ b9 ZAACA 369 300 FE V2034
70 ACBC~ 26 ~37 ( N . H h ) ~ CR.O~ 67 ZEKGA 12W 19~. FE
ACSAA 25 W~ RBZ CR2°O? V I l C0(2~)
~70 CJCHA k8 537 RB2 CRZ°07 V I I I 50 NATUA 165 kk2 CO 0
"1~71 ACSAA 25 35 RB2 CRZ,O7 X b8 CJCHA 4b 3't72 C03 04
71 JSSCB 3 36w AG2 CR, OW. 71 ACBCA 2 ; Zw3Z CO (C3 H6 03)
69 ACACB 25 $ 1 1 6 AG2 CRZ.O? 70 PEPll 2.61 C02 SI 3W
UNPU1 REF8 AGZ CR~ 07 70 PEPI1 3 161 BETA CWZ $1.OW
MN(2~} 62 ACCRA 15 1219 CO S°O~ (HZ.O)B
60 ACCRA 13 325 L I HN ~ 0 ~ 74 HCACA 5h. 1621 C03 (O.H) (S,OW) 2 ( H 2 , 0 )
70 NJHMA 113 1 MN? NAt2 ($.0k) 13 (HZ.O)15 ~ ACCRA 22 775 CO (NeHW] ~ (S.O~)Z ( H 2 ° O I ~
70 ZEKGA 132 1 MN5 [OoH) Z ( S I , OW}2 B~* ACCRA 17 2WO CO TI 33
67 AMMIA 52 709 CA HN $IZ 06 b9 ZAACA 369 30b CO V2 3W
63 ZEKGA 119 117 ZN2 HN (O.H)2 SI. OW UNPU% REF3 C02 V2 O?
70 ACBCA 26 6~0 CA2 MN (P,OW) Z ( H 2 , 0 ) Z UNPU1 R;F3 CO V2 06
65 ANMIA 50 188W *LAUEITE • MN FEZ ( O , H ) 2 ( P , O ~ ) 2 ( H 2 , 0 ) 8 UNPU1 REF3 C03 V2 08
69 AMHIA 5k 132 =PS-LAUEITE ~ MN FE2 ( P , O k ) 2 ( O , H ) 2 [H2,O (}8 ZEKGA 12b ;'99 CO GE 03
53 NJHMA MN S,O~. 70 CJCHA W8 881 C03 AS2 08
62 ZEKGA 117 331 CA MN SI2 06 70 CJCHA ~8 312L* C08 AS3 016
59 JUPSA 14 1352 MN T I 03 68 ZAACA 385 125 CO SE.OW
69 ZAACA 369 306 MN V2 Ok 65 ACCRA 19 269 CO H O 04
6? ACSAA 21 590 MN2 V2 07
 I. D . B R O W N AND R. D. S H A N N O N 281

DONNAY,G. (1969). Carnegie Inst. Wash. Year/). 68, 292- KHAN, A . A. & BAUR,W . H . (1972). A eta Cryst. B 28, 683-693.
295. LAZAREV,A. N. (1964). Izvest. Akad. Nauk SSSR, Set.
DONNAY, G. (1972). Acta Cryst. A28, S 108. Khim. pp. 235-241.
DONNAY, G. & ALLMAN, R.. (1970). Arner. Min. 55, 1003- LINNETT, J. W. (1940). Trans. Faraday Soc. 36, 1123-1134.
1015. MARTIN, R. F. & DONNAY, G. (1972). Amer. Min. 57, 554-
EVANS, H. T. JR (1960). Z. Kristallogr. 114, 257-277. 570.
FISHER, E. L. (1973). To be published. NOEL, W. (1963). Angew. Chem. Int. Ed. 2, 73-80.
FUMI, F. G. & Tosr, M. P. (1964). J. Phys. Chem. Solids, ORGEL, L. E. (1960). An Introduction to Transition Metal
25, 31-43. Chemistry: Ligand-field Theory. London: Methuen.
GARCIA-BLANCO, S. & FAYOS,J. (1968). Z. Kristallogr. 127, PANT, A. K. & CRUICKSHANK, D. W. J. (1967). Z. Kristal-
145-159. loAf. 125, 286-297.
GIEEESPIE, R. J. & NYHOLM, R. S. (1957). Quart. Rev. 11, PAUEING, L. (1927). J. Amer. Chem. Soc. 49, 765-790.
339-380. PAULING, L. (1929). J. Amer. Chem. Soc. 51, 1010-1026.
GOLDSCHMIDT, V. M., BARTH, T., LUNDE, G. & ZACHARIA- PAULING, L. (1940). The Nature of the Chemical Bond, 2nd
SEN, W. H. (1926). Skr. Norske Vidensk. -Akad. Oslo 1. ed., p. 72. Ithaca: Cornell Univ. Press.
Mat.-Nat. KI. No. 2. PAUEING, L. (1947). J. Amer. Chem. Soc. 69, 542-553.
GOPAE, R. (1972). P h . D . Thesis, McMaster Univ. p. PERLOFE,A. (1970). Inorg. Chem. 9, 2228-2239.
162. PREWITT, C. T. (1973). To be published.
HAMILTON, W. C. (1958). Phys. Rev. 110, 1050-1057. RIBBE, P. H. & GIBBS, G. V. (1969). Amer. Min. 54, 85-94.
ISHIKAWA, Y., SATO, S. & SYONO, Y. (1971). J. Phys. Soc. SHANNON,R. D. (1971). Chem. Commun. pp. 881-882.
Japan, 31,452-460. SHANNON, R. D. & CAEVO, C. (1973a). To be published.
JEFFREYS, G. A. (~ SLAUGHTER, M. (1963). Acta Cryst. 16, SHANNON, R. D. & CAEVO, C. (1973b). To be published.
177-184. SHANNON, R. D. & PREWITT, C. T. (1969). Acta Cryst. B25,
KAEM/~N,A. (1971). J. Chem. Soc. (A), pp. 1857-1859. 925-946.

Table 18 (cont.)
69 MSAPA 2 31 LI.FE SIZ 06 62 ZEKGA 117 3WW MG (N.,'iW) 2 (SCOW) 2 ( H 2 , 0 ) 6
71 SPHCA 15 993 FE3 82 S I 3 09 ( F j O . H ) 2 70 AMMIA 551535 °COQUIIBITE° (ALpFE)2(SoO~.)3 (H2o0)9
65
68
AMMIA
ACBCA
50
2W
1E27
1518
NAB S13 08
M ( AL3 SI B 09 ~ 66
70
ACCRA
ACBCA
21 383
25 1 6 9 6
CS AL ( S , OW) 2 H 2 , 0
K2 $ 5 . 0 1 6
67 ZEKGA 125 286 CA B S I . O ~ OeH 67 ZEKG, 12W 398 K2 CA (S. OW)2 H2.O
67 ACCqA 22 37 NA2 SI 03 ,; ANMIA 56191'
68 ACBCA 2~ 13 ALPHA ~A2 S I 2 05 65 AMMIA 50 1 9 2 9 [S.OW) 2 H2eO
68 ACBCA 2W 107/ BETA NA2 S I 2 05 71 AMMIA 56 1917 K FE ( $ o 0 k ) 2 .H2eO)w
66 AHHIA 51 950 NA AL ~12 Ob 68 ZEKGA 127 261 FE2 (S, Ok.)Z 0 ( H Z . O ) T
67 ZEKGA 125 298 CA2 NA H $13 09 70 ANHIA 55 78 • IOEME~ZTE • FE3 ( S ° O ~ ) k ' H 2 . O ) IW
70 ZEKGA 131 31~ CA NA AL S I 2 . 0 7 66 ACCRA 21 705 NI S,O~ [ D 2 , 0 ) 6
69 MSAPA 2 31 NA FE S I Z 06 68 ACBCA 2W 5 0 8 CU S , 0 ~ ( H ~ ° O ) 3
69 HSAPA 2 31 NA CR 312 06 69 ACBCA 25 19 CU [ N ° H 3 ) W S,pOW H 2 , 0
67 ACSAA 21 1W25 NA IN 31~ 06 69 ACBCA 25 676 CU (N~.HW) 8 (S.OW) 2 ' H Z . O ] 6
69 ZEKGA 129 65 MG S 1 0 3 69 ZE~GA 13o ~ , 9 (N.H~)8 $Z.08
71 AMMIA 56 115~ MG7 SI3 01~ F O,H 7 1 PHDTA PA R8 H S.OW
65 AMHIA 50 202~ MG3 ALZ S I 3 012 K(I+)
68 AMMIA 53 80~ MG FE SI,OW (THREE STRUCTURES} 70 ACBCA 26 1 1 8 9 K B 02
65 TMPMA 10 1 CA HG ~ I . OW 63 ACCRA 16 975 K2 BW 37 ( H 2 . 0 ) 4
69 NSAPA 2 31 CA MG 312 06 63 ACCRA 16 376 K 85 O~ [ H 2 , O ) L
61 ZEKGA 115 269 ~ANDALJSITE~ AL2 SI 05 71 INUCA 7 873 KK H C2 OW
63 ZEKGA 118 33? •KYANITE~ AL2 SZ O~ 53 PRLAA 220 39F H2 P. OW
63 ZEKGA 118 12T •SILIMANITE • AL2 SI 05 71 ACSAA 25 512 K H5 (D°OW)2
66 ZEKGA 123 81 CA3 ALZ SZ3 012 69 ZEKGA 130 1~8 ZR2 , P . OW) 3
70 ACBCA 20 233 SI P2 37 III 62 ACCRA 15 1280 L I K2 P3 09 H 2 . 0
70 PEPI1 3 161 BETA CAZ SI,OW 58 JACSA 62 925 BETA-K2 ScOW
71 ACBCA 27 8W8 GAMMA CA2 SI.OW 67 ZE~GA 12W 398 K2 CA ( S ° 0 4 ) 2 H2. O
68 AMMIA 53 80T F ,Z SI,OW 71 ANMIA 56 1 9 1 7 K FE (S,OW)2
67 CIWYA 67 285 FE SI 03 65 AMMZA 50 1 9 2 9 K FE (3°OW) Z H2.O
62 ZEKGA 117 331 ~BUSTA~ZTE = CA MN S I 2 06 71 AMMIA 56 1 9 1 7 K FE [S.OW)2 (HZeO)W
67 AMMZA 52 709 ~JOHANNSENITE e CA MN SZZ 06 6g ZEKGA 1 1 ~ 257 K V, 03
7t ZSTKA 12 272 Y82 SIZ 0 7 60 ZEKGA 11W 257 V , 0 3 HZ°O
70 ACBCA 26 W8W E~2 SI~ 0 7 66 IHOCA 5 96? K2 ZN2 VlO 0 2 8 ( H 2 . 0 ) 1 6
P(5+) 70 ACSAA 2k 3627 K2 CR3 08 OeH
55 ACSAA 9 1557 H3 P°O~ 69 J C S I A 1 9 6 9 10~1 KZ P8 (CU ( N ° O 2 ) b )
6 9 ACBCA 25 776 (H3 P . 3 ~ ] Z H 2 . 0 71 MOCHA 102 1 2 k 5 KZ GE~, 09
6 9 ZAACA 368 231 (H3 P. 3k] 2 H2.O 70 ACBCA 26 1W51 K2 SE Ok
60 ACCRA 15 325 LZ MN P.OW 69 JCSIA 1969 8W9 K2 NO Ok
6?INOCA 6 119 L I 3 P°3W 68 JCSIA 1908 1398 KZ M03 010
CJCHA REF2 NAW P2 O~ 66 ACSAA 20 2195 K2 M03 010
6 5 ACCRA 18 26 NA3 P 3 0 9 6W INOCA 3 63W K TE 0 ( O , H ) 5 H2, O
60 ACCRA 13 203 NA5 P3 010 66 ACSAA 20 2138 K~ TE2 06 (O,H)W [ H 2 ° O ] X
71 AC8CA 27 212W NA3 P, Ok ( H 2 . 0 ) l Z 65 INOCA 5 51~ KZ 8A GO ( H ° O 2 ) b
67 ACCRA 22 W3 NAw P 2 . 0 7 ( H 2 . 0 ) 1 0 69 ACBCA 25 170~ K2 W°OW
70 ACBCA 26 J58W NA2 H o . OW ( H 2 ~ 0 ) 7 CA(2~)
70 ACBCA 26 1W29 H ( N.HW P.OW . H 2 . 0 ) 6 6W ACCRA 17 685 CA 0
67 ACCRA 23 289 ALPHA :1G2 P2 07 57 JCPSA 26 56~ CA (O.M)Z
68 ACSAA 22 1 ~ 6 6 MG3 P2 08 66 ACCRA 20 2 9 5 CA12 BF..17 0 2 9
6 7 ACCRA 23 ~18 HG H P. OW ( H 2 . 0 ) 3 63 &CCRA 16 390 CA 82 O~-I
68 INOCA 7 13W5 ALPHA-CA2 PZ 07 69 ACBCA 25 955 CA 82 O ~ - I I I
66 ACCRA 21 9W2 8ETA-CA2 P2 07 69 NJHHA 1 9 6 9 1~2 CA3 82 36
70 INOCA 9 2259 CA2 P.O~ CL 70 ZEKGA 132 2W1 CA 63 05 (O,H)
6 6 ZEKGA 123 161 ALP.O* 68 NJMMA 1968 80 CA AL 3.0W
68 AMMZA 53 1096 AL2 P. 3W ( O . H ) 3 67 ZEKGA 125 286 CA B SI.OW 'O,H)
70 ACBCA 26 233 S I P2 07 57 ACCRA 10 567 ~CALCITE * CA C.03
6 9 AMHIA 5W 132 HN FE2 ( P . O ~ ) 2 , O . H ) Z ,HZeO)8 71 JNBAA 75 62779 ~ARAGO~ZTE 'D CA C . 0 3
53 PRLAA 229 39? K H2 P*OW 59 AMMIA ~W CA MG ( C ° 0 3 ) 2
~1 ACS~A .~ 512 ~ H5 ,~.o~)2 67 ZEKGA 125 298 CA2 NA H S I 3 09
6 9 ZEKGA 130 1~8 K ZR2 , P . OW)3 65 ZEKGA 122 175 CA NA2 ( S ° 0 ~ ) 2
70 ACBCA 26 1 8 2 6 .N.HW]~ P.OW ( H 2 . 0 ) 3 65 THPHA 10 1 CA MG S I , OW
JSSCB REFW V P 05 71 ACBCA 27 2 3 1 1 CA2 FE AL 05
67 ACCRA 22 665 ALPMA-CU2 P2 07 66 ZEKGA 123 81 CA3 AL2 S I 3 0 1 2
70 JSSCB 1 120 ALPHA-ZN~ P2 07 71 ACBCA 27 8k8 GAMMA-CA2 S I , 011
71 ZAACA 380 51 P CL5 62 ZEKGA 117 331 CA MN S I Z 06
S(6~) 67 AHMIA 5Z 7 0 7 CA MN S I 2 06
6 8 ACBCA 2W 2 9 9 HE S.OW H2. O 71 AMHIA 56 791 CA3 FE2 S I 3 0 1 2
6 9 JCPSA 51 W213 ( H 3 . 0 ) 2 SoOW 68 INOCA 7 13W5 ALPHA-CA P 2 . 0 7
65 ACCRA 18 717 L I S°O~ H 2 . 0 66 ACCRA 21 942 8 ETA-CA P 2 . 0 7
UNPU1 REF5 L I N Z . H 5 S.O~ 70 ACBCA 26 6WO CA2 MN ( P . O W ) 2 H 2 . 0 ) 2
67 ACCRA 22 906 L I NA S.OW 70 GHDCA 270 952 CA V2 O6
70 ACBCA 25 2 2 9 8 N.H~ L I S,OW 66 ACCRA 21 397 CA3 V13 0 2 8 (H2,0)17
al. 6 9 ACBCA 25 310 8E (HZ.O) W S.OW 70 ACBCA 26 1W69 CAZ FEZ 0 5
6 9 ACBCA 25 30~ BE ( H 2 . O ) ~ S . O k 71 NRBUA 6 725 8A CA FEW 08
70 ACBCA 26 335 ZN (C (N. H2) 3 ) 2 ( S . O ~ ) Z 65 ACCRA 18 88 CA W O~
65 ACSAA 1 9 1 6 1 2 [ N Z . HS) Z 5,04 SC(Z~)
70 ACBCA 26 536 H 2 . H 6 S,OW 67 ZEKGA 12~ 136 SCZ 03
67 ACCRA 22 182 NA AL (S. OW)2 ( H 2 . 0 ) 1 2 67 ARKEA 29 3W3 SC2 03
~ 6 5 ZEKGA 122 175 CA NA2 ( S . O W ) 2 65 ZAACA 339 130 L I SC 02
70 8SCFA 1970 4 2 ~ HG S.O~ H2°O 62 ACGRA 15 ~91 SC2 SIZ 0 7
6W ACCRA 17 863 MG S.O~ ( H 2 . O ) ~ 56 AGCRA 9 677 SC P.~O~
6~ ACCRA 17 235 MG S.OW ( H 2 . 0 ) 6 65 ZAACA 336 295 CA 5 C 2 0 W
6W ACCRA 17 1361 MG S.O~ ' H 2 . 0 ) 7 52 JPGHA 56 1W5 SC V,O~
65 ZEKGA 122 161 MG K2 ( 5 . 0 ~ ) 2 (H2°0)6 66 JCPSA h5 2 7 ~ 5 SC2 W3 0 1 2
282 EMPIRICAL BOND-STRENGTH-BOND-LENGTH CURVES FOR OXIDES

Table 18 (cont.)
AS(5*)
CO(Z*) 66 ACCRA 21 808 (AS2 05)3 (H2,0)5
70 ACBCA 76 8 CU O 70 ACSAA 24 3711 LI MO,OZ AS.Ok
71 ZAACA 379 15? L I E CU OZ 61 AMMIA 46 1077 CA2 B AS, Ok (O.H) W
58 ACCRA 11 169 CU8 MG8 ( C , 0 3 ] k ( O , H ) Z k ( H I , O ) 8 70 ACBCA 26 157W NA2 N AS. Ok (H2.0){
71 ACBCA 27 67? CU BZ 3k 70 ACBCA 26 158W NA2 H AS, Ok (HI.O) T
70 ACBCA 76 1203 (CU ( N . 0 3 ) 2 ) 7 ( H 7 , 0 ) 5 71 ACBCA 27 ZlZW NA3 AS,Ok (HZ,O)12
61 ACCRA tk 738 NA7 CU ( S , 0 ~ ) 7 ( H 7 , 0 ) 7 UNPU1 REF3 NAW AS2 07
71 AMMIA 56 193 CUE CA? SI3 010 (HZ.O)~ 70 CJCHA k8 890 MGZ AS2.37
67 ACCRA ZZ 665 ALPHA GUT P7 O? UNPUt REF9 HG17 AS6 032
68 CJCHA ~6 605 BETA CUE P2 07 70 ACBCA 26 1889 {N.H~)2 H AS.O~
63 ACCRA 16 1009 CUE 0 S,Ok 70 ACBCA Zb WOO CA H AS.Ok
63 NATUA 197 70 CU3 S,04 [O.M)4 69 ACBCA 75 15kw CA H AS.Ok (M2.0) 2
68 ACBCA 2~ 508 CU S,OW t H E , O ) 3 70 INOCA 9 2259 CA2 AS, Ok CL
62 PRLAA 266 95 CU S.O~ ( M 7 . 0 ) 5 70 CJCHA 48 881 C03 AS2 08
69 ACBCA Z5 676 CU (N.Hk) Z ( S , 0 ~ ) 7 (HZ°O)6 ?O CJCHA k8 312~ C08 AS] 016
UNPU1 REF9 CU V 03 I 68 AMMIA 53 T33 MN7 (O.M)8 (AS°Ok]2
UNPU1 REFI CU3 VZ 08 70 AMMIA 55 707] HN9 (O.H) 9 ( H 2 . 0 ) 7 AS,DO ( A S . 0 4 ) Z
UNPU1 REF% CU5 V2 010 70 AMMIA 55 1W89 HN7 SB AS 017
65 ACCRA 18 ??7 CU3 AS. Ok ( 0 , H ] 3
66 ACCRA 21 k3Z CUZ AS. Ok O.H (HI.O)3 65 ACCRA 18 77? CU3 AS. Ok (O.H)3
68 CJCHA 46 917 CU3 AS? 08 66 ACCRA 71 k3? CUZ AS. Ok O,H (H7°0)3
69 ZEKGA 130 231 CU ZN7 AS7 08 68 CJCHA h6 91l CU3 AS? 08
68 JCPSA k8 7619 CU HO Ok 69 ZEKGA 130 231 CU ZNZ AS? 08
71 ACBCA 27 2066 CU3 M07 09 70 ZEKGA 132 332 C03 AS? 08
70 ACBCA 20 1020 CU W O~ 71 AMHIA 56 11W? ZNW (AS.Ok)7 (O,H)Z ( H I , O ) 7
69 ACSAA 23 15 CU3 W 06 69 ACBCA 75 2658 ZR (H AS. Ok)7 HE°O
ZNt7~)
69 ACBCA 25 1233 ZN 0 I ~EFI R,D,SHANNON UNPUBLISHED
6Z SCIEA 137 993 ZN 0 I I REF2 K°Y.LEUNG AND C,CALVO TO BZ PUBLISHED
70 ZEKGA 131 213 ZN3 87 06 ~EF3 C,CALVO PRIVATE COMMUNICATION
71 ACBCA 27 677 ZN B~ 07 ~EF~ R. GOPAL AND C.CALVU TO BE PUBLISHED
70 ACBCA 26 335 ZN (C (N, H 2 ) 3 ) 7 (S,OW) 2 REF5 I,O,BROWN UNPUBLISHED
63 ACCRA 16 748 ZN (C. H3 C.O C,H C.O C,HO]2 H2.O REF6 H.N.NG AND C.CALVO TO BE PUBLISHED
71 ACBCA 77 12~0 2N (0 0 C C,O N.H2)2 [ H 7 . O ) 2 ~EF? BAYSIOE RESEARCH CENTER GENZRAL TELEPHONE AND ELECTRIC LAB, ABSTRACTS
61 ACCRA 22 ZkO ZN (O, HE) 6 (N ,O3) 7 REF8 R.G,HAZELL PRIVATE COMMUNICATION
TO ACBCA 26 860 I N 5 (OeH) 8 ( N . 0 3 ) 7 ( H 2 , 0 ) 2 ~ZF9 E,KOSTINER PRIVATE :OMMUNICATION
?i ACBCA Z? 1868 ZNZ SI.Ok ACTA CRYSTALLOGR, SECTION A
67 ZEKGA 17k 180 ZNW SIZ O? (O H]2 HT,0 ACACB ACTA CRYSTALLOGR. SE3TION B
67 ZEKGA 12k 115 PB ZN S! OW ' ACBCA
63 2EKGA 119 117 ZNZ MN ( O , H ) 7 SI. Ok ACCRA ACTA CRYSTALLOGR.
bb ZEKGA 123 3Z1 IN70,H P.OW ACSAA ACTA CHEM. SCANO,
70 JSSCB 1 120 ALPHA ZN7 PZ D? AM'IIA AMER,MIN,
65 CJCMA ~3 11kT B E T A ZNZ P2 O? ~CSJA BULL,CHEM. SOC.JAPAN
67 CJCHA 45 2303 BETA 2N3 P2 08 3SCFA BULL.SOC,CHIM,FR,
68 CJCHA k6 903 CO IN7 P2 08 3JFCA BULL,SOC.FR,MINER,CRISTALLOGR,
68 ZEKGA 126 27 Zh5 tO, H) 8 CL2 HE,0 CAMIA CANAO.HINER,
66 INOCA 5 95? K7 ZN7 ViO 028 KHT.O)16 C~OCA C,R, ACAO,SC.PARIS SECTION C
69 ZAACA 36~ 300 ZN VZ 3~ 3IWYA CARNEGIE INST,WASHZNGTON YEARBOOK
UNPU1 REF3 2N2 V2 07 3JCHA CANA3,J,CHEM,
71 CJCHA k9 3056 ALPHA ZN3 VZ 08 CRSCt CRYST.STRUCIURE COMH.
69 PHSSA 32 91 ZN FE7 Ok 3ZYPA CZECH°J.PHYS,
69 ZEKGA 130 Z3I CU ZNZ AS7 08 ~CACA HELVtCHIM, ACTA
71 AMMIA 56 11W? ZNW (AS.Ok]7 (O,H)2 (H7°0)2 I~OCA INORG,CHEM,
67 JCPSA W6 2052 ALPHA ZN NO Ok JACSA J,AMER,CHEM,SOC.
66 ACCRA 21 k87 ZNZ MOI 08 JCPSA J,CHEM.PHYS.
GA(3¢) J3SIA J.CHEM.SOC°LOND, SECTION A
67 JCPSA k6 1862 ALPHA CA7 03 JESOA J,ELECTROCHEH.SCC.
60 JCPSA 33 6T6 BETA CA2 03 JI~CA J,INORG,NUCL,CHEM,
65 ACCRA 18 k81 L ] CA OZ JNBAA J,RES.NAT,BUREAU STAND, SECTION A
65 JCPSA 42 3957 GA{2-X) FE(X) 03 JPCHA J,RHYS,CHEM,
68 JCPSA ;8 1[9~ EU3 FEZ CA3 012 JPCSA J.BHYS,CHEM,SOLIDS
65 ACCRA 19 971 Y3 CA5 312 JSSCB J,SOLIO STATE CHEM.
65 ACCRA 18 87w CA NB Ok JUPSA J.PHYS,SOC, JAPAN
68 JIMCA 30 1389 IN GA 03 ~IASA MINER.MAC,
65 ACCRA 19 971 YB3 GAS 017 MOCHA MDNATSHEFTE CHEM.
65 ACCRA 19 971 LU3 GAS 017 ~RBUA MATERIALS RESEARCH BJLLETIN
GEt&~) ~SAPA MIN, SOC,AMERICA SP{C,PAPERS
64 ACCRA 17 842 GE.O7 ~ATUA NATURE LONDON
69 MOCHA 100 79 L I NA GEW.09 NJ~HA NEUES JAHRBUCH MINER, MONATSHEFTE
71 MOCHA 107 96~ NtZ GE,03 =Eall PHYS,EARTH PLANETARY INTERI3RS
63 ACSAA 17 61Y NAW GE~ 020 DHSSA PHYS,STAT,SOLIDI
&? ACCRA 22 507 NA7 GE 03 (H~°O)b PRLAA PROC,ROY.SOC,LONOON SERIES A
&CSAA ZW NAW SNZ GEW 017 (O,H)k SCIEA SCIENCE
SOVIET PMYS, CRYSTALLOGRAPHY (SEE KRISA]
JSSCB 617 MGZ8 G£10 Ok8 ~SCOA SOLIO STATE COMMUNICATIONS
71 MOCHA 102 12~5 KZ GE~,09 TFSOA TRANS.FARAOAY SOC,
69 ZEKGA 130 139 MN GE 03 TMaHA TSCHERMAKS MINER,PETR,MITT,
7 0 PEPII 3 161 BETA-MN2 GE.Ok UNPU1 (UNPUBLISHED REFERENCE)
69 ZEKGA 129 ~27 MN3 FE~ GE3 01~ ZAACA Z,ANORG.ALLGEM.CHEH°
68 ZEKGA 126 299 CO GE 03 ZEKGA Z. KRISTALLOGR,
?O SPHCA 15 36 E~7 GEZ 3? ZSTKA ZH. STRUKT,KHIM,

SLAUGHTER, M. (1966). Geochim. Cosmochim. Acta, 30, TEMPLETON, D. H. (1953). J. Chem. Phys. 21, 2097-2098.
299-313. ZACHARIASEN,W. H. (1931). Z. Kristallogr. 80, 137-153.
SMITH, J. V. & BAILEY, S. W. (1963). Acta Cryst. 16, 801- ZACHARIASEN,W. H. (1954). Acta Cryst. 7, 795-799.
811. ZACHARIASEN,W. H. (1963). Acta Cryst. 16, 385-389.
SON, P. R. & BARTELS, R. A. (1972). J. Phys. Chem. Solids, ZACHARIASEN, W. H. • PLETTINGER, H. A. (1959). Acta
33, 819-828. Crvst. 12, 526-530.

View publication stats