An earth scientist’s periodic table of the elements and their ions

L. Bruce Railsback*
Department of Geology, University of Georgia, Athens, Georgia 30602-2501, USA

ABSTRACT
A new Earth Scientist’s Periodic Table of the Elements and Their Ions presents the
naturally occurring charged species commonly encountered by geoscientists, as well as
elemental forms, and it is organized by charge. The new table therefore shows many
elements multiple times, unlike the conventional table. As a result, trends, patterns, and
interrelationships in mineralogy, soil and sediment geochemistry, igneous petrology, aque-
ous geochemistry, isotope geochemistry, and nutrient chemistry become apparent in this
new table. The new table thus provides a more effective framework for understanding
geochemistry than the conventional, and purely elemental, periodic table.

Keywords: geochemistry, mantle, minerals, nutrients, seawater, weathering.

INTRODUCTION DESCRIPTION OF THE TABLE

The Periodic Table of the Elements formulated by de Chancour-
tois, Meyer, and Mendeleev (Farber, 1969; Courtney, 1999) has clearly
been of great utility in explaining and predicting relationships in chem-
istry. It has been of less utility, however, in the earth sciences. For
example, it does not arrange lithophile, siderophile, and chalcophile
elements into distinct groups, and it does not group elements into nat-
urally occurring sets (e.g., elements concentrated in the mantle, in sea-
water, or in soil). Elements critical for biological processes are likewise
not grouped in useful ways by the conventional periodic table. In these
respects, the conventional periodic table has not provided a good
framework for understanding the chemistry of Earth and its life.
Application of the conventional periodic table of the elements to
the earth sciences has been disadvantaged because most matter at or
near Earth’s surface is not in elemental form. Instead, most atoms of
the matter encountered by earth scientists carry charge. Si is a very
good example: every earth scientist has encountered Si as Si41, where-
as few earth scientists are even aware that a very small amount of
natural elemental Si is known to exist (Gaines et al., 1997). The use-
fulness of any document summarizing chemistry for the earth sciences
would clearly be enhanced by inclusion of charged matter in addition
to elemental forms.
With that view in mind, this paper presents an Earth Scientist’s
Periodic Table of the Elements and Their Ions1. In this table, natural
groupings and trends in geochemistry, marine chemistry, and nutrient
chemistry become apparent, allowing a more general synthesis of the
chemistry of the earth sciences. The result is an integrated view of
geochemistry applicable from the mantle to soil to seawater. One fun-
damental concept in rationalizing these geochemical patterns is the dif-
ference in bonding exhibited by hard and soft cations, which favor O22
and S22, respectively. The other is the extent to which charge of cations
is sufficiently focused (i.e., ionic potential is sufficiently high) to pro-
vide strong bonds to O22 without causing repulsion between those
cations. Thus bonding and coordination with oxygen, Earth’s most
abundant element in the mantle and crust (McDonough and Sun, 1995),
dictate many of the trends discussed in the following sections.
*E-mail: rlsbk@gly.uga.edu.
1GSA Data Repository item 2003109, sources of information used in con-
Chemical entities are arranged according to charge in the Earth
Scientist’s Periodic Table of the Elements and Their Ions (Fig. 1)2.
Thus, B, C, and N, which are conventionally on the right side of the
periodic table, appear on the left in the left-to-right horizontal sequence
Li1, Be21, B31, C41, and N51. Al, Si, P, and S similarly appear on the
left and within the left-to-right horizontal sequence Na1, Mg21, Al31,
Si41, P51, and S61. A more striking result of organizing the table ac-
cording to charge is that many elements appear multiple times, because
different natural conditions cause those elements to assume different
charges. Many elements (e.g., P and U) thus appear twice, a few appear
three times (e.g., V, Fe, C, and N), and a few appear four times (most
notably S, as S22, S0, S41, and S61).
The table is broken from left to right to separate noble gases, hard
or type A cations (those with no outer-shell electrons), intermediate to
soft or type B cations (those with at least some outer-shell cations),
elemental (uncharged) forms, anions, and the noble gases again (Fig.
1). The significance of the division of cations is that hard cations bond
strongly to F2 and O22 but not to S22, whereas the soft cations bond
strongly to S22 and the larger halides, Br2 and I2 (Stumm and Morgan,
1996) (Fig. 1, Inset 8). These patterns are exemplified in nature by the
absence of sulfide minerals of Ca21 and of the other hard cations, but
the existence of oxides and sulfates of those cations (Fig. 1). The nat-
ural occurrence of sulfides, but not oxides, of the platinum group ions
provides a converse example (Fig. 1). The differences between hard
and soft cations are further illustrated by insets 3 and 6, which show
that melting temperatures of oxides of cations of intermediate ionic
potential decrease from hard to intermediate to soft cations. Inset 8
likewise shows that the relative solubility of halide compounds can be
predicted from the division of hard and soft cations, in that solubility
of halides of hard cations increases from F2 to I2, whereas solubility
of halides of soft cations increases from I2 to F2. In igneous geochem-
istry, the failure of Cu1 to bond with O22 and substitute for Na1 in
plagioclase (Ringwood, 1955) provides an example of the different
behavior of soft and hard cations, respectively. The failure of Tl1 to
substitute for K1, despite the similar size and charge of those two
cations, is another example.
Another difference between the new table and its conventional
predecessors is that the new table includes the naturally occurring ac-
2Loose insert: Figure 1. An Earth Scientist’s Periodic Table of the Ele-

structing table, and explanatory notes, is available online at www.geosociety.org/ ments and Their Ions. Sources of information used in constructing table and
pubs/ft2003.htm, or on request from editing@geosociety.org or Documents Secre- related notes are available in Appendix DR1 in GSA Data Repository (see text
tary, GSA, P.O. Box 9140, Boulder, CO 80301-9140, USA. footnote one).

q 2003 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org.
Geology; September 2003; v. 31; no. 9; p. 737–740; 1 figure; Data Repository item 2003109. 737
 Ions commonly concentrated in residual soils and residual
Outline solid for naturally occurring elements or ions; sediments. Small symbol ( ) indicates less certainty. 4 most abundant constituents
dashed for ones that rarely or never occur in nature. Ions concentrated in deep-sea ferromanganese nodules in atmosphere
Elements that occur as native minerals, recognized in antiquity
( recognized from Middle Ages to 1862; Fe 10 most abundant elements in Earth's crust
5th to 8th most abundant recognized after 1963.) Zr 11th to 20th most abundant elements in Earth's crust
Ions least depleted from mantle in formation of crust
Symbol
Cations that
"Hard" or "Type A" Cations (see scale at far right) Atomic Number Ions that enter early-forming phases in igneous rocks Cations that form simple fluoride minerals
Elements that make natural mineral alloys with Fe Li 21st to 40th most abundant elements in Earth's crust
coordinate with H2O (number of protons)
Ions that enter later phases in igneous rocks because of Cations that form simple oxide minerals Elements that make natural mineral alloys with Cu Lu 41st to 92nd most abundant elements in Earth's crust
(or CO32– or SO42–) (All electrons removed from outer shell)
(Thus a noble-gas-like configuration
Element Name
Ge
3+
54 Ionic Radius (r) (Å)
their large size (mostly "large-ion lithophiles") Cations that form simple sulfide minerals Elements that make natural mineral alloys with Os Noble Gases
in solution Atomic Mass Actinium
8 most abundant solutes dissolved in seawater Cations that form simple bromide or
(No ionization)
of the outer shell) (or elemental radius Elements that make natural mineral alloys with Pt
m=72.59
H+
for elemental forms) iodide minerals Elements that make natural mineral alloys with Au
1 Coordinate F>O>N=Cl>Br>I>S r=1.05 9th to 16th most abundant 17th to 22nd most abundant
Cations that form oxysalt minerals –
He 2
Naturally Helium
Most abundant (bold) Anions H 1

2
Hydrogen ion
Cations that See also Insets 1 to 6. Most abundant solute in average river water (HCO 3–) (e.g., S6+ in sulfates, As5+ in arsenates) Elements that are thought to make up most of the

r =
Hydrogen
occurring 234 Radioactive (italicized) Earth's core (Fe>Ni>Co), along with possibly S or O as hydride m=4.0026
m=1.0079 coordinate 2nd to 8th most abundant solutes in average river water See also Inset 7.

z
/
isotopes EC, `+ Anions that form minerals with Ag+ m=1.0079 r=1.2
Noble Gases r=10-5 with OH– (or Cations that Cations that coordinate with `-
H2O) in O2– in solution (e.g., as
_
Solutes that can be limiting nutrients in the oceans Anions that form minerals with Au+ Anions that commonly coordinate with H+ r=2.08
(No ionization) z coordinate with OH–
/ =4 123 NO3–, PO43–, SO42–, etc.) Radioactive z = ionic charge ÷ Anions that form minerals with K+ and Na+ Gases (e.g., as CH4, NH3, H2S, H2O, etc.) 123 34
z/ r = solution or O2– in solution r ionic radius Macronutrient solutes on land Micronutrient solutes on land
r 2 + decay pathways Anions that form minerals with Mg2+ –
He 2 LiLithium 3 Be2+ 3+
4 B Boron 5 C 4+
6 NNitrogen 7
5+ = ionic potential 6 7 8 C 4– 3–
6 N 2–
7 O 8 F asFluorine 9 10

which hard cations

Helium
or charge density Ions essential to the nutrition of at least some vertebrates Anions that form minerals with Al3+, Ti4+, and Zr4+ Non- Ne
ion Beryllium ion as borate (B(OH) Carbon, as CO , ("essential minerals") C N2 O2 Reduced carbon Reduced nitrogen

preferentially
2
metals Oxygen as oxide fluoride Neon

Anions with
bicarbonate (HCO-3) as nitrate (NO3– )
3
m=4.0026 m=6.941 Anions that form minerals with Si4+

coordinate
m=9.012 or B(OH)4– )
& carbonate (CO32- ) Diamond Molecular Molecular m=12.011 m=14.007 m=15.999 m=18.998 m=20.180
r=1.2 r=0.60 r=0.31 m=10.811 m=14.007 & graphite nitrogen oxygen r=2.60 r=1.71 r=1.36
r=0.20 m=12.011 r=0.11 r=1.40 r=1.5
z = 32 = ionic charge ÷
r=0.77 r=0.71 Most natural occurrences of carbides and
r=0.15 r ionic radius z/ = nitrides are in meteorites or mantle phases.
34 67 9 10 11 12 13 14 14 15 r 16 Metals 12 13 14 14 15 16 17 18 19 20 21 22
Cations that
Na+ 11 Mg2+12 Al 13 Si 4+ –
Ne Neon 10 3+ 5+
14 P 15 SSulfur as 16
6+
Intermediate Cations coordinate with O2– (± OH–) in solution Elemental Forms 13 14 16 4–
Si 14 P3– 15 S2– 16 Cl 17 Ar 18
Aluminum ion as as silicate (SiO44– ) 4+ 16 5+ 6+ S
m=20.180
z
/r =
Sodium ion Magnesium ion
Al3+ or Al(OH)3–n or Si(OH)04
Phosphorus as
2–
(Some electrons remain in outer shell) S Sulfur As 33 Se 34 (uncharged) Al Si Silicon as silicide Phosphorus Sulfur as sulfide
Chlorine
as choride
Argon

Intermediate
phosphate (PO43– sulfate (SO4 ) as phosphide
1 m=22.990 m=24.305 n
z as sulfite (SO 2–)
arsenate (AsO43–) as selenate (SeO42–) Principal elements in iron Sulfur m=39.948
r=1.5 r=0.95 r=0.65 m=26.982 m=28.086 and HPO42– ) m=32.066 Coordination with S or O likely z
/r = 8 r=0.37
3
other than noble gases Aluminum Silicon m=28.086 m=30.974 m=32.066 m=35.453

16
r=0.41 m=30.974 r=0.47 r=0.42 meteorites (Fe>>Ni>>Co) and, r=1.43 r=1.34 r=2.71 r=2.12 r=1.84 r=1.81 r=1.8
r=0.50 r=0.29 /r =
3+
24 Mn3,4+ 25 Fe3+ 26 Co3+ 27 Ni 3+ 28 Cu2+ 29 Sn4+

=
r=0.34 CrChromic 4 50 Sb5+ 51 Te6+ 52
with S or O, presumably domi- The only known natural occurrences of

z
/r
4+
20 21 22 Where Fe2+
and Fe3+ would 23 24 25 26
Fe 3+ 27
28 29 30 31 32 33 34 36 3+
23
VVanadium ion Manganese ion
chromium Ferric iron Cobaltic cobalt Nickel ion Cupric copper
Stannic tin
r=0.71 e.g., as antimonate e.g., as tellurate
nant elements in Earth's core phosphides and silicides are in meteorites
and cosmic dust. 32 33 34 36 35 37 36 38 40
Fe 2+

18
fall if they were
K+ 19 Ca 2+ 20 3+ 4+ 5+ 6+ Ti 22 r=0.61 m=51.996 4+ r=0.53 2+
30 Ga 3+
31 Ge4+ 32
r=0.62 r=0.56
24 26 27 28 29 30 33 34 2–
34 –
35 36
Ar hard cations
Sc 21 Ti 22 V 23 Cr 24 Titanium ion
3+
23
r=0.69 3+ r= 0.64 r=0.64 r=0.63 r=0.73 r=0.69 Zn Gallium ion Germanium ion 3+
33 Se4+ 34 As3– 33 Se Br Kr
Argon Potassium ion Calcium ion Scandium ion Titanic titanium Vanadium ion Chromium, e.g. as r=0.75 VVanadous
50 52 53 54 Mn 25
2+ 2+
27 Ni Nickel ion28 Cu 2+
29 m=65.39 2+ + Zinc ion
m=69.723 m=72.61
As
Arsenic Selenium Cr Fe Co Ni Cu Zn (Atomic masses As Se Arsenic as arsenide
Selenium
as selenide
Bromine Krypton
m=39.098 m=40.078 m=44.956 m=47.867 e.g., as vanadate chromate (CrO4 2–)
2+ 26 Co Fe e.g., as selenite and isotopic as bromide m=83.80

which soft cations

vanadium Nickel Copper
m=39.948 e.g., as arsenite Chromium Iron Cobalt Zinc Arsenic Selenium m=78.96
r=1.33 r=0.99 r=0.81 r=0.68 m=50.942 m=51.996 Ti 22 m=50.942 Manganous Mn Ferrous iron Cobaltous cobalt
m=58.693 Cuprous copper
r=0.74 r=0.62 r=0.53 m=74.922 m=78.96 r=1.26 r=1.25 r=1.24 r=1.28 r=1.39 information m=74.922 m=79.904 r=1.9

preferentially
r=1.8 2+
24 m=54.938 m=55.845 m=58.933 r=1.27 r=1.48 r=1.6

Anions with
CrChromous m=63.546 (2+ r=0.93) r=0.50 are omitted to r=2.22 r=1.95

coordinate
Titanium ion r=0.74 (1+ r=1.13) r=1.98
r=0.59 r=0.52 r=0.74 r=0.72 64 66 r=0.69 conserve space)
40 42 43 45 r=0.90 r=0.80 r=0.76 58 60 r=0.96 70 72 74 76 77 (7+ r=0.39)
78 80 82
46 47 48 50 51
chromium
67 68 70 69 71 75 74 76 77
36 38 40 39 40 41 44 46 48 ? (48) 49 50 50 51 50 52 53 54 r=0.90 55 54 56 57 58 59 61 62 64 63 65 73 74 76 78 80 82 75 78 80 82 79 81 (82) 83 84 86
4+ 2+ 1,3+ 3+ 4+ 5+
Kr 36 2+
Rb+ 37 Sr 38 Y3+ 39 Zr 40 Nb 41 Mo 6+ 42
4+ 5+ z
/r
Mo 42 Tc
Molybdenum ion
43 Ru3,4+44 Rh2+ 45 Pd2+ 46 Ag + 47 Cd 48 In 49 Sn
2+
50 Sb 51 Te 52 I Iodine 53 44 45 46 47 48 49 50 51 52 Sb
3–
51 Te 2– 52 I
–
53 Xe 54
Krypton = Technetium Ruthenium ion Rhodium ion Cadmium ion Indium ion Antimony ion, Tellurium ion,
as iodate (IO3– ) Tellurium Iodine as iodide Xenon
Rubidium ion Strontium ion Yttrium ion Zirconium ion Molybdenum Niobium (or 8 m=95.94 Very limited m=101.07 m=102.906 Palladium ion m=107.868 m=112.411
Silver ion
m=114.818 Stannous tin as in antimonites as in tellurites Ru Rh Pd Ag Cd In Sn Sb Te Antimony as telluride
m=126.904 m=131.29
m=118.710 m=121.760 m=127.60 m=126.904
m=87.62 e.g., as molybdate
m=83.80 m=85.468 m=88.906 m=91.224 Columbium) ion
natural m=106.42 3+ r=0.81 Ruthenium Rhodium Palladium Silver Cadmium Indium Tin Antimony Tellurium as antimonide
m=127.60
r=1.13 r=0.93 r=0.80 m=92.906 m=95.94 r=0.68 occurrence 3+ r=0.69 r=0.86 r=1.26 r=0.97 r=0.44
r=0.89 Iodine m=121.760 r=2.21 r=2.16 r=2.1
r=1.9 r=1.48 on Earth 4+ r=0.67 r=0.86 1+ r=1.32 r=1.12 r=0.90 is shown twice r=1.34 r=1.34 r=1.37 r=1.44 r=1.56 r=1.66 r=1.58 r=1.61 r=1.7 (7+ r=0.50)
r=0.70 r=0.62 96 98 99 106 108 110 112 114 115 116 120 122 123 as a solute in seawater r=2.45 120 122 123 124 126 128
84 86 90 91 92 94 95 96 99 102 104 105 111 112 113 because it speciates
129 130 131
78 80 82 92 94 95 97 (100) 100 101
106 108 110 107 109 113 115 117 118 119
121 123 124 125 126 both as I– (to right)
124 125 126 (124) 127
83 84 86 85 87 87 88 89 92 94 96 ? 93 (96) 96 98 100 97 98 100 102 104 103 114 116 120 122 124 128 130 and IO 3– (here). 121 123 (128) (130) 132 134 136
128 130
+ 3+
Xe 54 Cs+ 55 Ba 2+
56 La3+& 57- Hf
4+
72 5+
73 6+
74 Re7+
W
4+
74 Re 75 Os 76 Ir 77 Pt 78 Au 79 Hg+ 80
4+ 4+ 4+ 2+ + Tl 81 Pb2+ 82 Bi 83 Po 84 73 75 76 77 78 79 80 81 82 83 Bi
2–
83 At 85 Rn 86
Ta W 75 Tungsten (Wolfram) Iridium ion Thallous thallium Bismuth ion

1
Xenon Rhenium ion Osmium ion Platinum ion Gold ion Mercurous ion Plumbous lead Polonium
REEs 71

2
Ir Tl Pb

z/ r=
m=192.217 m=195.078 m=196.967 m=200.59 m=204.383 m=207.2 m=208.980
Barium ion Hafnium ion Tungsten (Wolfram) Bismuth as
Re Au Hg Radon

r =
Cesium ion Tantalum ion Rhenium ion ion Ta Os Pt Bi Astatine
m=183.84 m=186.207 m=190.23
bismuthide
m=131.29 m=132.905 m=137.327 m=178.49 e.g., as tantalate e.g., as tungstate
r=1.40 r=1.20 (222)

z
/
m=180.948 m=183.84 m=186.207 r=1.20 Tantalum Osmium Iridium Platinum Gold Mercury Thallium Lead Bismuth
r=2.1 r=1.35 r=0.65 r=0.69 r=0.66 r=0.96 r=1.37 r=1.19
Rhenium m=208.980
r=1.69 r=0.64

8
r=0.81 r=0.56 r=1.46 r=1.37 r=1.35 r=1.35 r=1.38 r=1.44 r=1.60 r=1.71 r=1.75 r=1.82
124 126 128 130 132 See below r=0.73 r=0.68
180 182 183 184 186 190 192 193? (3+ r=0.85) 196 198 199 203 205 206 204 206 207 209 210 211 210 211 212 218 219
129 130 131 174 176 177 194 195 200 201 214 215 The only bismuthide
215 218 219 220 222
134 135 136 180 182 183 187 188 189 191 193 208 210 211 212 214 215 =1 minerals are of
132 134 136 133 137 138 170Yb 178 179 180 180 181 184 186 185 187 184 186 185 187 190 192 196 198 197 202 204 206 207 208 210 212 214 216 218 z /r Pd, Ag, Pt, Au, and Pb

4+ 2+ 3+ 4+ 5+
Fr +
4+
90 6+ Elements 95 and beyond do not occur naturally:
92 95: Americium 101: Mendelevium Hg 80 Thallic thallium
Tl 81 PbPlumbic82 Bi 83

1
86 2+ Th
87 Ac 3+ 89 Pa5+ 92 U

2
Rn 88 91 93 Pu 94

1
Ra U Np

/r =
Mercuric ion lead Bismuth ion

/r =

r/ =
Thorium ion Uranium ion
Radon Francium ion Actinium ion Protactinium ion Uranium, e.g. Neptunium Plutonium 96: Curium 102: Nobelium r=1.10 r=0.95 r=0.84 r=0.74

z
Radium ion

z
r=0.97

z
(222) (223) (226) m=227.03 ? m=232.038 (231) as uranyl (UO22+) Very limited Very limited
97:Berkelium 103: Lawrencium
2
98 Californium 104: Rutherfordium
r=1.18 r=0.95 m=238.029 natural natural =4
= 4
r=1.76 r=1.40 (+3 r=1.14) (+4 r=0.98) occurrence occurrence
99: Einsteinium 105: Hahnium
z /r z /r
z/r
=
r=0.7 100: Fermium

219 220 222

very rare
(<30 g in crust) 223 224 227 228 230 234 235*238
on Earth on Earth "Soft" ("Type B") Cations
223 226 228
227 228

*For the sake of simplicity,

231232 *234
231 234 237 ?

z =
239 (Many electrons remain in outer shell)
Coordinate I>Br>S>Cl=N>O>F An Earth Scientist's Periodic Table of the Elements and Their Ions
z/ = 1 =

the 235U-207Pb and /r 8

r

232Th-208Pb series are omitted.

z
Inset 1: Bulk modulus (Ks in GPa) /r = 4+
Ce 58 Rare earth elements (REEs)
ion ioni

4 3+ Cation
of oxide minerals of hard cations Eu 63 (effectively "Hard" or "Type A" cations in their 3+ state) Inset 4: Solubility of oxide minerals of hard cations
ic c c ra

Inset 2: Hardness of oxide minerals of hard cations Inset 3: High-temperature behavior of hard cations
z/r = 2

Cerium ion (outlined for

Mineral of Mineral of Europium ion intermediate
3+
har diu

r=1.01 3+ 3+ 3+
one cation: two cations: Nonmineral: 3+ Pr 3+ 59 Nd3+ 60 Pm 61 Sm 62 m=151.964 Gd 3+ 64 3+ 66 3+ 68 3+ Yb 70 3+ +
Be92+ C 4+ N 5+ Ti 4+ cations) + 2+
C 4+ N 5+
38 La 57 65 Dy
67 Er
71 Li B 3+ +
Be 2+ B 3+ C 4+ N 5+ Li4.4 Be–7.4 3+
B2.77
ge s

210 71 3+ Praseodymium ion

58 m=140.908 m=144.24 No natural m=150.36 151 153 m=157.25 Promethium Samarium ion r=1.03 Tb Ho Tm 69 Ytterbium ion Lu Li Melting T (K)
Lantha- Lanthanum ion Ce
Quartz Perovskite Neodymium ion Gadolinium ion Dysprosium ion Erbium ion 2103
÷

Cerium ion m=162.50 Holmium ion m=167.26 Thulium ion m=173.04 Lutetium ion Terbium ion Bromellite 1700 2681 723 En 216 of oxides of Bromellite
+ 2+ 3+ 4+ nides: m=138.906 m=140.116 r=1.09 r=1.04 m=158.925 m=164.930 m=168.934 r=0.94 m=174.967 Chrysoberyl Ab hard cations
Li Be B C occurrence

H=
r=1.08 2+ r=1.02 r=0.99 r=0.96 8.5 500 Minerals are shown with size
r=1.15 Eu 63 152 154 155 r=1.00 r=0.97 r=0.95 (2+ r= 1.13) r=0.93 Ab
2+Fo

6
251 r=1.11 on Earth 2+ 2+
Bromellite 3 (4+ r=0.92) 142 143 144 144 147 148 156 158
162 164 166 168 170 171 Na
+ Mg Spinel 3+ Al Si 4+
P 5+
S 6+ + Mg Al 3+ 4+ 5+ 6+
of circles representing +
Na9.9 Mg Al3+ Si 4+ 5+
P–1.37 S 6+
Chrysoberyl 136 138 140 146 145 ? 149 150 Europium ion
156 157 160 161 162 165 167 168 170 169 172 173 175 176 176Hf
7.5-8 9
Na 3125 2345 Si Q P S proportions of cations.
–2.4 –8.1 –3.9
240 138Ba 138 139 142 141 148 150 ? (150) 152 154 158 160 159 163 164 174 176 ? Periclase
5.5-6
7 1193 An 1996 855 290 Periclase
+ 2+Spinel 3+ 4+ 5+ r=1.12 Corundum Quartz Minerals in order of typical Corundum Quartz
Na Mg Al Si P z/r = 2 Substitutes for Ca2+
Fe
2+
Fe
3+ temperature of formation*:

H=
198 254 38 ----- Il

Ti-rich minerals
Mg-Al-Fe-Ca-
Periclase Quartz Ksp ----- Chromite
160 50

8
Corundum + Perovskite +
Inset 5: Typical simple oxysalt minerals Inset 8: Solubility of halides K Ca3.5 2+
5.5
Ti >9 4+
V 5+
Cr 6+
K
+ Di An
2+
Cr
Cr
3+
Ti 4+ V 5+ Cr 6+
Forsterite
Anorthite K14.0 Ca2+ Sc3+ Ti 4+ 5+
V –7.6 Cr 6+
(__MOn minerals without OH or H2O) Inset 6: Melting and decomposition (d) temperatures Inset 7: Conceptual model of the behavior of of hard and soft cations Lime * 3-3.5 Ca -----
Lime 1.4 –9.7

H=
5+ Augite
(Ru=6-6.5) Ti 2103 943 Rutile
+ 2+ Perovskite 4+ 5+ (K) of oxides of intermediate and soft cations 00
Cassiterite 4+ As
oxides of hard (and intermediate) cations Anion:
Shcherbinaite Bi 3200
Sc 3+ Enstatite
Shcherbinaite

4
K Ca Ti216 V 3+ 4+ 5+ 16 Sn 588
F–
Mineral Villiaumite
Srilankite 6.5 10 Hornblende
B C N Tenorite Sellaite

H=
210 3+ Hematite 0 1903 N (NaF) AgF 00
Lime 115 Rutile Mn 3+ 3+ 20 20002+ Li
Cations
Nonmineral (MgF2) Ilmenite
NaNO3 Co 1 Cu

4
MgAlBO4 Me2+CO3 (Soda Niter) Fe Argutite Arsenolite Mineral of 2+ 4+ + 2+
Nb5+ Mo6+ Zr 4+ Nb5+ Mo6+
Tausonite
Y 3+ Y 3+
Paramont- Eskolaite 1353(d) +
H 1Å Magnetite
Zincite
Sr + Rb Sr4.3
175 200 (Sinhalite) (e.g.,
Calcite)
KNO3
Minerals with
1+ cations only
roseite 4+
V Cr 3+
2+
1838 1168 (d)
2+ Bunsenite
1719
Cuprite Zn
2+ Ga
3+
Ge
4+
As
3+
Cl– Chlorargyrite HgCl2 Halite one cation: Zr 6.5
3-4 Rb Sr 2+ Y 3+ Zr 4+ Nb5+ Mo6+ Apatite
28.9
(Nitre) 2603 Fe 2+ 2+ 2079 547 (NaCl) 7 Baddeleyite 673 2938 3123 1785 1074 Titanite (sphene) Baddeleyite
2240 Mn Co Ni Cu + 2242 1388 (AgCl) Molybdite Molybdite

rich minerals
+ 2+ 3+ 4+ 5+ 2400 2054 1652 Quartz Hardness Zircon
Rb Sr Y 1 Zr Nb 2078 2228 1509 MgCl2
Si 4+ P 5+ S 6+ Manga- Wüstite Monteponite Romarchite Valentinite (Mohs Biotite

Si-Na-K-
87 50 Bromargyrite
152*
KAl2Si3O8 (Kspar) Na3PO4 CaSO4
2000 Mo4+ nosite 1600 Rh
3+
Pd
2+
Ag+
2+ 3+
Sn
Cd
2+ 3+
In Sb
High z/r Br– (AgBr) HgBr2 MgBr2 NaBr Mineral of scale) K-feldspars

La 3+ Hf 4+ Ta 5+ W 6+
two cations
La 3+ Hf 4+ Ta 5+ W 6+ 2+
*Baddeleyite has
10

1373(d) 1373(d) 1023(d) ~473(d) >1773 1353(d) Strong + Albite

Ba2+ La 3+ Hf 4+ Ta 5+ W 6+
(Olympite) (Anhydrite) 2185
Ks = 95 GPa but Al2 SiO5 (K-S-A) Tugarinovite
928
O2– bonds, but 5.5 Cs Quartz Ba6.7
0

Rb
0

AlPO4 Na2SO4

1500
Iodargyrite

H=6
is not the most
(Thenardite) Minerals I– 7
120

stable ZrO2 phase

ZrSiO4 (Zircon) (Berlinite)
with Au3+ Hg + Tl + Massicot Bismite Intermediate cation-cation (AgI) HgI2 MgI2 NaI Perovskite Tantite
2286 2580 3173 2058 1745 *Order of crystallization in Mineral Tantite
+ Low z/r
0

2+ 3+ 4+

2000
repulsion any one magma depends on

1500
423(d)
80

z/r

200

2500
373(d)
Cs Ba La Hf

2500
at ambient condi- 1+ & 2+ 4+ 4+ 3+ 2+ 852 2+ 3+ bulk composition, pressure,
"K-S-A" Minerals W Rh Ir Pt Pb Bi Weak cation- Strong cation- 10-8 10-6 10-4 –9.7
0

tions; value shown 2+ 10-2 1 100

Th 5+ Th 5+
cations + and fluid composition.

3000

300
40

71 indicates with Au Hg 3+ 1170

Th 5+
145 1273 (d) 598(d)

0
is for the latter. kyanite, Minerals with ~1773(d) 1173(d)
no stable 773(d) Tl 1098 oxygen bonds oxygen bonds Solubility of Ag+( )-, Hg2+( )-, *A non-rutile synthetic TiO2
Melting T (K) of simple Log of activity of cation species
andalusite, 1+ to 4+ cations 1+ to 3+ oxide Montroydite1107
Na+( )-, and Mg2+( )-bearing halides (mol/L)
is the hardest known oxide Thorianite 6 Thorianite

0
cations Avicennite 3493 in distilled water at 25 °C
& sillimanite. See also Inset 3. See also Inset 6. oxides of hard cations
v. 4.6

Figure 1. An Earth Scientist’s Periodic Table of the Elements and Their Ions. An earth scientist’s periodic table of the elements and their ions
Sources of information used in constructing table and related notes are available in L. Bruce Railsback
Appendix DR1 in GSA Data Repository (see text footnote one). Figure 1
Supplement to Geology, v. 31, no. 9 (September 2003)
tinides with the hard cations. Th 41 thus falls below Hf41, and U61 falls
below W61. Trends in the symbols described in the next section extend
across this unconventional but geochemically useful arrangement (Fig.
1).
The table also shows the atomic numbers, atomic masses, natu-
rally occurring isotopes, and naturally occurring decay paths of the
different elements. Names of elements and ionic forms are shown—
e.g., ‘‘Sulfur as sulfate (SO242)’’ for S61. Sizes of chemical symbols are
scaled to abundance of the elements in Earth’s crust; seven of the nine
most abundant elements conveniently fall together in one part of the
left side of the new table. Contours of equal ionic potential (charge 4
radius, i.e., z/r) highlighted in blue and brown extend across the table
and parallel trends in natural occurrences, as discussed in the next
section.
PATTERNS AND TRENDS IN THE TABLE
Symbols in the Earth Scientist’s Periodic Table of the Elements
and Their Ions show natural occurrences or enrichments in minerals,
natural waters, soils and sediments, igneous rocks, the mantle, and the
atmosphere, and as critical nutrients (Fig. 1). These symbols fall in
swaths that follow contours of equal ionic potential across the table,
as one would expect from Cartledge (1928a, 1928b), Goldschmidt
(1937), and Mason (1958). As a result, the new table makes apparent
patterns of geochemistry that do not emerge from the conventional
table. For example, many ions with ionic potential between 3 and 10
make oxide minerals, are concentrated in soil and ferromanganese nod-
ules, enter early-forming igneous phases, and are least depleted the
mantle. The result is a red-and-brown swath across the hard and inter-
mediate cations in the new table (Fig. 1). The same swath of hard
cations includes those that make oxides with the highest melting tem-
perature, lowest solubility, greatest hardness, and greatest bulk modu-
lus (insets 1–4 and 6).
On the other hand, hard cations with ionic potential ,4 make
fluoride minerals, include ions abundant in river water and seawater,
and include ions important as nutrients. Cations with ionic potential
.8 likewise include ions abundant in seawater, ions important as nu-
trients, and ions that form oxysalts, such as sulfates and arsenates. The
results are blue-and-green swaths across the new table. Those swaths
extend from the hard cations (which coincide with the lithophile ele-
ments) to the intermediate to soft cations (which coincide as a whole
with the siderophile and chalcophile elements).
Contours of ionic potential continuing from the hard cations to
the intermediate cations (e.g., Mn41 and Fe31) continue the red-and-
brown swath across the table, in that intermediate cations with ionic
potential of 3–8 also make oxide minerals, are concentrated in soils,
enter early-forming igneous phases, and so on. On the other hand, the
contours for lowest ionic potential (1–2) set off the soft cations, which
include the coinage metals (Cu, Ag, and Au) and form the center of a
region characterized by yellow diamonds that mark ions forming sul-
fide, bromide, and iodide minerals.
The coinage metals and their neighbors are also shown in a section
of the table highlighting elemental forms (the true ‘‘Table of the Ele-
ments’’ within the new table). Symbols and colored fields show that
groups of these elements make alloys. For example, elements alloying
with Os form a small distinct group, and elements alloying with Fe
form a group overlapping little with elements alloying with Cu and
Au.
On the right side of the table, patterns among the anions match
those on the left side. Among the anions of low ionic potential, a blue-
and-green swath of symbols pertaining to solutes mirrors that found in
cations of low ionic potential. From top to bottom is the transition
from anions coordinating with hard cations (resulting in fluorides and
oxides of Na1, K1, and Al31) to those coordinating with soft cations
738
(resulting in sulfides, bromides, iodides, and tellurides of Ag1 and
Au1). From right to left, or from Cl2 and F2 to O22 to C42, is the
transition from anions making minerals with hard cations of low ionic
potential (e.g., K1 and Na1) to those making minerals with hard cations
of higher ionic potential (e.g., Al31 and Si41). Minerals exemplifying
this transition are carrobite (KF), sellaite (MgF2), gibbsite (Al2O3),
quartz (SiO2), and moissanite (SiC).
Speciation of hard cations in aqueous solution also follows easily
recognized trends in the new table (bold black lines in Fig. 1). From
lower left to upper right across the hard cations, speciation progresses
from hydration (e.g., K1) to hydroxo complexes [e.g., Al(OH)3–n n and
Si(OH)04] to oxo-hydroxo complexes (e.g., COOOH2 and POOOOH2,
more familiar as HCO23 and HPO242) to oxo complexes (e.g., CO232,
NO23 , and SO422) (Stumm and Morgan, 1996; Shock et al., 1997). A
trend that is inscrutable in the conventional periodic table thus becomes
readily apparent in the new table. This pattern of coordination, when
extended from solution to solids, places in context the mineralogical
existence of oxysalt minerals such as carbonates, nitrates, and sulfates,
but no ‘‘calciates,’’ or ‘‘sodiates’’, and the existence of aluminate only
as an aqueous species (Pokrovskii and Helgeson, 1997).
Many of these patterns can be explained by consideration of bond
strength and shielding of charge in mineral structures, as is shown
schematically in inset 7. Cations of low ionic potential (e.g., K1, Na1,
Sr21) bond only weakly to O22, so they do not form oxide minerals
and are not retained in oxide-forming and hydroxide-forming environ-
ments like soils. Instead, they are soluble in aqueous solution, entering
natural waters and crossing cell membranes and root sheaths as nutri-
ents. Their weak bonds to O22 result in their incorporation into igneous
minerals only at relatively low temperatures and thus relatively late in
the crystallization sequence. Cations of intermediate ionic potential
(e.g., Al31, Ti41) form relatively strong bonds with O22, and their tet-
rahedral to cubic coordination allows shielding of the cations’ positive
charges from each other. They thus form stable oxides and hydroxides
in oxidizing environments, and many of them bond in igneous minerals
at high temperatures and thus early in the crystallization sequence.
Their stability as oxides and hydroxides results in low solubility and
low concentration in natural waters and thus leads to their irrelevance
as nutrients. Cations of high ionic potential (e.g., P51, N51, S61) form
very strong bonds with O22 in radicals like PO342 , NO23 , and SO422, but
their intense concentration of incompletely shielded positive charge and
resultant repulsion preclude formation of oxides or hydroxide minerals.
Thus, like cations of low ionic potential, they are soluble in aqueous
solution, abundant in natural waters, and cross cell membranes and
root sheaths as nutrients. Their concentration of positive charge causes
them to enter igneous minerals so late that they are among the ‘‘in-
compatible’’ ions in crystallization of silicate magmas.
These considerations help explain the existence, and nonexistence,
of oxysalt minerals (inset 6). Simple silicates (silicates without OH2
and/or H2O) built around Si41 can accommodate 11 to 41 cations.
Simple borates and phosphates, built around cations of higher ionic
potential (B31 and P51), can only accommodate 11 to 31 cations,
presumably because residual positive charge from borate and phosphate
groups repels 41 cations. Simple carbonates and sulfates, built around
cations of even higher ionic potential, accommodate only 11 and 21
cations. Finally, simple nitrates, built around the tiny highly charged
N51, only accommodate 11 cations, presumably because the unshield-
ed positive charge from nitrate groups repels any cations of 21 or
greater charge. The same trends, with shifted thresholds, exist in anal-
ogous minerals with OH2 and/or H2O (e.g., hydrous nitrates accom-
modate 11 and 21 but not more highly charged ones, and OH-bearing
sulfates and carbonates accommodate 11 to 31 but not 41 cations).
The result is a predictive model of the existence and nonexistence of
oxysalt minerals of various cations.
GEOLOGY, September 2003
GENERAL INSIGHTS
Perhaps the most general insight apparent in the new table is that
chemical weathering at Earth’s surface and the evolution of Earth to
separate the mantle and continental crust are geochemically much the
same process: the segregation of hard and intermediate cations of low
and high ionic potential from those of intermediate ionic potential. The
result is (1) concentration of many of the cations of intermediate ionic
potential in the mantle and, at Earth’s surface, in soil (the red-and-
brown swaths in the table) and (2) the ultimate removal of ions of low
and high ionic potential to the oceans (blue swaths in the table). The
process continues in the oceans, in that cations of intermediate ionic
potential are segregated to ferromanganese nodules and have a short
residence time in seawater.
Another major Earth process—life—has followed rules similar to
those dictating mantle evolution and weathering. Because life began
and largely evolved in aqueous solution and because chemical entities
must be dissolved to pass through cell membranes, life has utilized
and depends on soluble chemical forms. The critical nutrients for life
(green symbols in the table) are therefore coincident with the chemical
species dissolved in natural waters (blue symbols in the table). The
evolutionary transition of some life forms to land put them in an en-
vironment in which chemical weathering removes such ions from soils.
The result is a conundrum for water-loving plant life: soils rich in
nutrients are most common in arid regions where those nutrients have
not been removed by weathering, and soils where wet conditions favor
life are typically leached of nutrients. Utilization of 61 ions by both
plants and animals exemplifies this evolutionary challenge: modern
farmers commonly must fertilize plant growth with K1-bearing fertiliz-
ers, vertebrates frequent salt licks for Na1, and premodern societies trad-
ed NaCl as a precious substance. Modern humans continue the trend, in
that they consume I2-supplemented NaCl, drink F2-supplemented water,
consume K1-bearing sports drinks, and even take Li1 pills.
SIX EXAMPLES ACROSS THE PERIODIC TABLE
Special Nature of Silicon
In addition to the trends already outlined, many important special
cases in geochemistry become apparent with the new table. For ex-
ample, Si41 is unique in being very abundant (it is the second most
abundant constituent in the crust) and in having an ionic potential at
the boundary between the relatively insoluble cations of intermediate
ionic potential (e.g., Al31, Ti41, and Sc31 in the red-and-brown swath
of the table) and cations of high ionic potential that form soluble rad-
icals (e.g., C41, N51, P51, and S61 in the blue-and-green swath). Si41
is thus abundant both in residua from weathering (e.g., in sands and
sandy or kaolinitic soils) and in natural waters, such as river water
(where dissolved silica is the second most abundant dissolved species)
and seawater (where it is the 11th most abundant dissolved species).
The abundance and borderline ionic potential of Si41 also have
important implications in igneous petrology. Most igneous minerals are
silicates, but some of the first phases to form in igneous rocks (e.g.,
spinel and chromite) contain no Si at all, and the first Si-bearing min-
erals to form are forsterite (where Mg is more abundant than Si) and
anorthite (where Ca and Al outnumber Si) (inset 3). Only in later-
forming phases does Si41 become the dominant cation, and only at the
end of Bowen’s reaction series (when incompatible elements enter sol-
ids) does SiO2 form as quartz (inset 3). This paradox of Si41 as a
somewhat incompatible ion in the crystallization of silicate magmas
arises because Si41 is at the upper margin of ionic potentials that allow
formation of stable oxides. In fact, Si41 is just a step away from C41,
N51, and P51, which do not make any such oxide at all because of
their high ionic potential and are ‘‘incompatible’’ in igneous petrology.
The abundance and borderline ionic potential of Si41 also lead to
an interesting feature of plant physiology. Plants take in nutrients like
GEOLOGY, September 2003
NO23 as solutes, and the borderline ionic potential of Si41 lets it be
taken up as a solute [as H4SiO04, or more accurately as Si(OH)04]. How-
ever, Si41 is sufficiently insoluble that some plants build masses of
opaline silica in their tissue (Meunier and Colin, 2001). These mineral
accumulations within plants, called phytoliths, exist because Si41 is
sufficiently abundant and soluble to be taken up through roots in so-
lution but sufficiently insoluble to be maintained as a solid mineral
within wet plant tissue.
Extreme Nature of Gold
If Si has a special role in geochemistry because of its abundance
and location at a threshold in ionic potential, Au is special for opposite
reasons: it is extremely rare, it has an extremely low ionic potential as
Au1, and Au1 is extremely ‘‘soft’’ in the spectrum from hard to soft
cations. As a result, Au forms no oxide minerals and doesn’t even form
a sulfide of only Au; the only analogues of a hypothetical AuS2 are
two AuTe2 minerals and an Au2Bi mineral. Au’s fondness for large
anions is also seen in its extensive complexing with Cl2 in solution,
to which is attributed much of the transport of gold to generate ore
deposits (e.g., Murphy et al., 2000). On the other hand, the indifference
of Au to oxygen is one of the reasons Au has been so valued by
humans—it continues to shine as an unoxidized metal despite centuries
of exposure to O2. In the conventional periodic table, Au is simply one
of many elements in the middle of that table, but isolation by contours
of ionic potential illustrates its unique nature in the new periodic table.
Applications of Fe, Mn, and Ce as Paleoredox Indicators
The table’s separation of cations and consideration of ionic po-
tential shows why some cations have been used extensively in evalu-
ation of paleoredox conditions. For example, Fe and Mn have been
used in geochemical study of oxidation and reduction (e.g., Hem, 1972)
because they are by far the most abundant elements that form nonhard
cations. They are thus the most abundant elements that can undergo
changes of one in oxidation state (e.g., between 21 and 31 for Fe).
In oxidizing conditions, they are highly charged (5 31) small ions
that are insoluble because of their high ionic potential and resultant
formation of hydroxides. In reducing conditions, they are lesser
charged (21) and larger ions that thus have lower ionic potential and
are soluble. Fe therefore behaves like the soil-forming and oxide-
forming Al31 ion when oxidized to Fe31, but like the ‘‘weatherable’’
and soluble Mg21 ion when reduced to Fe21.
Ce is less abundant than Fe and Mn, but otherwise analogous in
that it precipitates in solids when oxidized to Ce41 but is more soluble
as Ce31. It thus provides an indicator of oxygenation in modern (de
Baar et al., 1988) and ancient (Wright et al., 1987) oceans. Ce31 and
Ce41 also exemplify patterns of coordination seen elsewhere in the
table, in that the ion with lower ionic potential forms a fluoride mineral,
fluocerite-(Ce), whereas the ion with higher ionic potential forms an
oxide mineral, cerianite. The formation of those two minerals parallels
the formation of fluorides by Na1 and K1 but formation of oxides by
Al31 and Sc31 (Fig. 1).
Uranium, Thorium, and Dating Problems
Radiometric dating using the U-series method (Edwards et al.,
1987) is a valuable means of determining the age of materials younger
than ;500 k.y. old. One problem with this method, however, is that
U at Earth’s surface is in the U61 state and thus in the soluble UO221
oxo complex (Langmuir, 1978). Th, on the other hand, occurs as Th41,
which is insoluble (e.g., Kaufman, 1969). As a result, parent U is
commonly lost from materials but daughter Th remains, giving incor-
rectly old ages (e.g., Dabous and Osmond, 2000). This problem is not
apparent from the conventional periodic table but is predictable in the
new table, where U61 falls in the blue-and-green swath of relatively
soluble hard cations of high ionic potential and Th41 falls in the red-
739
and-brown swath of relatively insoluble ions of intermediate ionic
potential.
From Silicates to Selenites
As already discussed, Si41 is located at the boundary between
cations of intermediate and high ionic potential in the new table. One
result is that Si41 is one of the few cations that make both an oxide
mineral (quartz) and oxysalt minerals (the many silicates). V51 and
Mo61 are hard cations with about the same ionic potential, and they
also form oxides (shcherbinaite and molybdite) and oxysalt minerals
(the vanadates and molybdates). If one follows the contour for z/r 5
8 from those hard cations to the intermediate and soft cations (Fig. 1),
one arrives at Se41, which likewise forms an oxide mineral (downeyite)
and oxysalt minerals (the selenites). Like Si41, V51, and Mo61, Se41
is also a cation essential to vertebrate nutrition (McDowell, 1992; Sun-
de, 1997). These similarities illustrate the continuity of trends along
contours of equal ionic potential across the table. The principal differ-
ence in the behavior of Se is that it forms cations with multiple outer-
shell electrons and thus exists as Se61 as well as Se41. Se61 also forms
oxysalt minerals (the selenates) and is important in nutrition, but, just
as one would expect after following the contour for z/r 5 16 from P51
and S61 in the hard cations (Fig. 1), Se61 does not form an oxide
mineral.
Odd Role of Chloride—Jack of All Trades, Master of None
Cl2 is not as abundant as F2 in the crust, but Cl2 is more abundant
than F2 in most natural waters. The reason for this paradox can be
seen in the new table. Cl2 occupies an intermediate position among
the anions at the right of the table, in that it coordinates with hard
cations to make soluble minerals like sylvite and halite and coordinates
with soft cations to make relatively rare minerals like chlorargyrite
(AgCl) (Fig. 1). By comparison, F2 bonds with hard cations well
enough to make insoluble minerals like fluorite and so is sequestered
in them, leaving relatively low concentrations in natural waters. Cl2,
in contrast, bonds strongly with neither hard nor soft cations and so
makes only relatively soluble minerals (inset 8). It therefore can reach
high concentrations in natural waters and commonly only precipitates
when it and a weakly bonding cation, Na1, finally achieve saturation
with respect to halite.
SUMMARY
Organizing the new Earth Scientist’s Periodic Table of the Ele-
ments and Their Ions according to charge yields an arrangement more
conducive to recognizing geochemical trends than that of the conven-
tional periodic table. These trends in mineralogy, aqueous geochem-
istry, igneous petrology, mantle geochemistry, soil and sediment chem-
istry, and nutrient chemistry are largely controlled by coordination of
cations with O22. This synthesis of geochemistry from mantle to soil
to seawater provides a framework for understanding Earth systems and
predicting geochemical relationships that is not recognizable with con-
ventional, elementally constructed, periodic tables.
ACKNOWLEDGMENTS
I thank my colleagues at the University of Georgia for their helpful com-
ments and especially Michael F. Roden for his insights about igneous processes.
The manuscript was improved by the comments of two anonymous Geology
reviewers and editor Ben van der Pluijm. Publication was supported by National
Science Foundation grant DUE 02-03115.
740
REFERENCES CITED
Cartledge, G.H., 1928a, Studies on the periodic system: I. The ionic potential
as a periodic function: American Chemical Society Journal, v. 50,
p. 2855–2863.
Cartledge, G.H., 1928b, Studies on the periodic system: II. The ionic potential
and related properties: American Chemical Society Journal, v. 50,
p. 2863–2872.
Courtney, A., 1999, A brief history of the development of the periodic table:
http://www.wou.edu/las/physci/ch412/perhist.htm.
Dabous, A.A., and Osmond, J.K., 2000, U/Th study of speleothems from the
Wadi Sannur Cavern, Eastern Desert of Egypt: Carbonates and Evaporites,
v. 15, p. 1–6.
de Baar, H.J.W., German, C.R., Elderfield, H., and van Gaans, P., 1988, Rare
earth element distributions in anoxic waters of the Cariaco Trench: Geo-
chimica et Cosmochimica Acta, v. 52, p. 1203–1219.
Edwards, R.L., Chen, J.H., and Wasserburg, G.J., 1987, 238U-234U-230Th-232Th
systematics and the precise measurement of time over the past 500,000
years: Earth and Planetary Science Letters, v. 81, p. 175–192.
Farber, E., 1969, The evolution of chemistry: New York, Ronald Press, 437 p.
Gaines, R.W., Skinner, H.C.W., Foord, E.E., Mason, R., and Rosenzweig, A.,
1997, Dana’s new mineralogy (eighth edition): New York, John Wiley and
Sons, 1819 p.
Goldschmidt, V.M., 1937, The principles of distribution of chemical elements
in minerals and rocks: Journal of the Chemical Society, p. 655–673.
Hem, J.D., 1972, Chemical factors that influence the availability of iron and
manganese in aqueous solutions: Geological Society of America Bulletin,
v. 83, p. 443–450.
Kaufman, A., 1969, The Th-232 concentration of surface ocean water: Geo-
chimica et Cosmochimica Acta, v. 33, p. 717–724.
Langmuir, D., 1978, Uranium solution-mineral equilibria at low temperatures
with applications to sedimentary ore deposits: Geochimica et Cosmochim-
ica Acta, v. 42, p. 547–570.
Mason, B., 1958, Principles of geochemistry: New York, John Wiley and Sons,
310 p.
McDonough, W.F., and Sun, S.-s., 1995, The composition of the Earth: Chem-
ical Geology, v. 120, p. 233–253.
McDowell, L.R., 1992, Minerals in animal and human nutrition: San Diego,
Academic Press, 524 p.
Meunier, J.D., and Colin, F., eds., 2001, Phytoliths: Applications in earth sci-
ences and human history: Exton, Pennsylvania, A.A. Balkema Publishers,
378 p.
Murphy, P.J., Stevens, G., and LaGrange, M.S., 2000, The effects of temperature
and pressure on gold-chloride speciation in hydrothermal fluids: A Raman
spectroscopic study: Geochimica et Cosmochimica Acta, v. 64,
p. 479–494.
Pokrovskii, V.A., and Helgeson, H.C., 1997, Thermodynamic properties of
aqueous species and the solubilities of minerals at high pressures and
temperatures; the system Al2O3-H2O-KOH: Chemical Geology, v. 137,
p. 221–242.
Ringwood, A.E., 1955, The principles covering trace element distribution during
magmatic crystallization: Part I. The influence of electronegativity: Geo-
chimica et Cosmochimica Acta, v. 7, p. 189–202.
Shock, E.L., Sassani, D.C., Willis, M., and Sverjensky, D.A., 1997, Inorganic
species in geologic fluids; correlations among standard molal thermody-
namic properties of aqueous ions and hydroxide complexes: Geochimica
et Cosmochimica Acta, v. 61, p. 907–950.
Stumm, W., and Morgan, J.J., 1996, Aquatic chemistry (third edition): New
York, John Wiley and Sons, 1022 p.
Sunde, R.A., 1997, Selenium, in O’Dell, B.L., eds., and Sunde, R.A., eds.,
Handbook of nutritionally essential mineral elements: New York, Marcel
Dekker, Inc., p. 493–556.
Wright, J., Schrader, H., and Holser, W.T., 1987, Paleoredox variations in ancient
oceans recorded by rare earth elements in fossil apatite: Geochimica et
Cosmochimica Acta, v. 51, p. 631–644.
Manuscript received 27 January 2003
Revised manuscript received 2 May 2003
Manuscript accepted 6 May 2003
Printed in USA
GEOLOGY, September 2003