p952.p65

Information • Textbooks • Media • Resources

952

Journal of Chemical Education • V ol. 80 No. 8 August 2003 • JChemEd.chem.wisc.edu

One of the few facts that I can remember from my un-

dergraduate inorganic course was my instructor’s insistence
that zinc, cadmium, and mercury should be classified as main-
block elements rather than as transition-block or d-block el-
ements. Though I have always assumed that the evidence for
this statement was unambiguous, I have also noticed the ap-
pearance over the last decade of an increasing number of gen-
eral chemistry texts, inorganic texts, and advanced inorganic
monographs that either explicitly or implicitly contradict this
assignment. The inorganic textbook by Cotton and
Wilkinson, which has served as the American standard for
nearly 40 years, has always been firm in its treatment of the
members of the Zn group as main-block elements, whereas
the text by Holleman and Wiberg, which has served as the
German standard in this field for nearly a century, has al-
ways classified them as outer-transition metals (1, 2). Like-
wise, the recent monograph by Massey on the main-block
elements treats them as members of the main block, whereas
the recent monograph by Jones on the d- and f-block ele-
ments treats them as transition metals (3, 4). A further ex-
amination of advanced monographs on coordination
chemistry and organometallic chemistry also revealed the
same inconsistency, with roughly a 50–50 split between those
texts that included the Zn group among the transition met-
als versus those that did not (5, 6).

In contrast to these explicit claims, many general chem-

istry and lower-level inorganic texts are ambiguous in their
treatment of these elements, often classifying them as transi-
tion metals in one part of the text and as main-block ele-
ments in another (7, 8). Much of this schizophrenic behavior
is attributable to the use of a periodic table similar to that
given Figure 1, which is designed to introduce students to
the names used for the various subdivisions of the table. I
have found a table of this type in virtually every recent gen-
eral chemistry text that I have examined, as well as in about

a quarter of the more recent introductory inorganic texts. In
all cases, the Zn group was incorrectly labeled as being a mem-
ber of the d block or transition block. Those introductory
inorganic texts that presented some sort of systematic survey
of descriptive chemistry usually contradicted this assignment
in their later discussions of the chemistry of these elements.
On the other hand, the surveys of descriptive chemistry found
in most of the general chemistry texts were so superficial that
the existence of this inconsistency seldom became explicit.

In light of these trends, I thought it might be of interest

to summarize the evidence relating to the proper placement
of the Zn group within the periodic table. This evidence can
be subdivided into three categories based on its relevance to
answering each of the following three questions:

•

Where does the transition block end?

•

Where does the d block end?

•

Where does the Zn group fit into the main block?

As we will see, past opinions on this issue have gener-

ally been a function of the particular form of the periodic
table advocated by the author in question, and for this rea-
son it is also of interest to review some of the various schemes
that have been proposed for graphically representing the
placement of the Zn group within the table itself.

What Is a Transition Element?

During the last quarter of the 19th century, the term

“Übergangsmetalle” or “transitional metal” was generally used
to describe the metal triads of Fe-Co-Ni, Ru-Rh-Pd, and Os-
Ir-Pt found in group VIII of the periodic table. Prior to the
discovery of the noble gases, the maximum oxidation states
of the known elements were found to undergo a cyclic varia-
tion from one to seven (Newland’s law of octaves). Each short
period of the table corresponded to one such cycle (e.g., Na
to Cl), whereas each long period corresponded to two such
cycles (e.g., K to Mn and Cu to Br). These latter cycles were
separated by a triad of metals (e.g., Fe-Co-Ni) whose prop-
erties corresponded to a gradual transition between those of
the last member of the first cycle (e.g., Mn) and those of the
first member of the second cycle (e.g., Cu)—hence the name
“transitional” or “transition” metal. With the discovery of the
noble gases, it was further suggested that they also be classi-
fied as transitional elements, since they, like the noble met-
als, also bridged the gap between one valence cycle, Li–F, and
the next, Na–Cl (9–11).

The first use of the term “transition” in its modern elec-

tronic sense appears to be due to the British chemist C. R.
Bury, who first used the term in his 1921 paper on the elec-
tronic structure of atoms and the periodic table (12). As early
as 1916, G. N. Lewis had suggested that the unique proper-
ties of the elements in the center of the longer (i.e., 18-ele-
ment) periods might be due to the presence of “variable”
kernels or cores in their atoms, though he did not elaborate

The Place of Zinc, Cadmium, and Mercury in the Periodic Table

William B. Jensen

Department of Chemistry, University of Cincinnati, Cincinnati, OH 45221-0172;

jensenwb@email.uc.edu

Figure 1.

A commonly used periodic table that incorrectly places

the Zn group in the transition or d block.

Information • Textbooks • Media • Resources

JChemEd.chem.wisc.edu • V ol. 80 No. 8 August 2003 • Jour nal of Chemical Education

953

on his suggestion (13). Using Lewis’ hint as a starting point,
Bury postulated that the outermost or n shell of a noble gas
atom (other than He) always contained eight electrons, but
once that shell became part of the atomic core for the atoms
of the succeeding period, it could, beginning with period 4,
expand its electron occupancy from 8 to 18 in the case of
the n

−

1 shell, or from 18 to 32 in the case of the n

−

shell. Bury referred to these as the “8–18 transition series”
and the “18–32 transition series”, respectively, and used them
to rationalize the electronic structures of both our current d-
block and f-block elements. Bury’s particular use of the term
“transition” was presumably intended to indicate that these
elements were undergoing a transition in the occupancy of
their underlying n

−

1 or n

−

2 shells from 8 or 18 electrons

at the beginning of the series to 18 or 32 electrons at the
end of the series.

The eventual adoption of Bury’s terminology by chem-

ists is probably due to G. N. Lewis, who summarized Bury’s
conclusions in his 1923 monograph on Valence and the Struc-
ture of Atoms and Molecules, where he explicitly refers to the
elements of Bury’s series as “transition elements” (14). Simi-
lar suggestions concerning shell-filling sequences were made
simultaneously by Niels Bohr, based largely on spectroscopic
rather than chemical evidence (15). However, to the best of
my knowledge, Bohr never used the term “transition element”
in his writings (16).

Though the work of Bury and Bohr was soon incorpo-

rated into textbook discussions of the electronic structure of
the atom, most texts remained conservative in their coverage
of the periodic table, and it was not until the end of the Sec-
ond World War that the older identification of the group VIII
metal triads with the term “transition metal” was completely
displaced by the newer electronic usage (17). As we will see,
it is not an accident that this change also coincided with a
widespread switch from the older “short” 8-column version
of the periodic table (Figure 2) to the so-called “long” or 18-
column version of the table (Figure 3). By the early 1940s,

Figure 2. An updated version of the short or 8-column block form
of the periodic table showing the original meaning of the A and
B subgroups or series.

the terms “transition” and “inner transition” had also largely
replaced Bury’s original numerical modifiers, and the term
“representative element” had become a common synonym for
the members of the main block (18, 19).

Where Does the Transition Block End?

In his famous review article of 1871, Mendeleev hesi-

tated between classifying Cu, Ag, and Au with the alkali met-
als in group I or with the transition metals in group VIII,
thereby converting them into transitional-metal tetrads (Table
1, line 1). In the end, he listed them in both locations, but
enclosed them in parentheses in order to indicate the tenta-
tive nature of their assignments (20). Group assignment in
Mendeleev’s table was based on maximum valence or oxida-
tion state and the existence of Cu(II) and Au(III) compounds
definitely precluded the assignment of these elements to
group I. Yet, if they were excluded from group I, a funda-
mental asymmetry appeared in the short form of periodic
table. Groups II–VII all bifurcated after period 3 into what
Lothar Meyer later called A and B “Untergruppen” or sub-
groups. If the metals of the copper group were not forced
into group I, then it would fail to show a similar bifurca-
tion. Mendeleev was at least honest enough to indicate his
indecision but, in the case of his successors, Platonic sym-
metry soon triumphed over the facts of chemistry. Thus, in

Figure 3. An updated version of the 18-column block form of
the periodic table showing various labeling systems for the
groups.

Information • Textbooks • Media • Resources

954

Journal of Chemical Education • V ol. 80 No. 8 August 2003 • JChemEd.chem.wisc.edu

most pre-electronic periodic tables, the transition elements
began at Fe, Ru, and Os and ended at Ni, Pd, and Pt, rather
than at Cu, Ag, and Au (Table 1, line 2).

In his book (14 ), G. N. Lewis suggested that the transi-

tion elements could be chemically differentiated from the
main-block elements not only by characteristic differences
in their overall patterns of observed oxidation states, but also
by their ability, in at least some of these states, to form col-
ored, paramagnetic ions (Table 2). To Lewis’ original list, we
can also add the characteristic absence of stereoactive lone
pairs on the central atom in their lower oxidation-state com-
pounds (21). Of these four characteristic properties, Bury
focused almost exclusively on oxidation-state patterns as a
criterion for identifying which elements qualified as transi-
tion metals. All active valence electrons (

) were assumed to

be in the outermost n shell. When an element exhibited an
oxidation state less than the number of electrons (

) added

since the preceding noble gas, this indicated that the excess
electrons were no longer acting as valence electrons but had
gone into the n

−

1 or n

−

2 shells instead. Thus the exist-

ence of one or more oxidations states for which

was

taken as evidence that the atom in question was a transition
element.

Based on the oxidation states known at the time, Bury

found that the first transition series began at Ti and ended
at Cu, that the second transition series began at Ru and ended
at Pd, and that the third transition series began at Os and
ended at Au (Table 1, line 3). As can be seen, Bury’s crite-
rion is highly dependent on the state of our knowledge of
the known oxidation states of the elements. As our knowl-
edge of the latter increases, so will the number of transition
elements. If one updates Bury’s assignments using a modern
inorganic text, his criterion would classify all elements from

the Sc group through the Cu group as transition metals (Table
1, line 5).

Using similarities in the arc spectra of the elements as a

way of characterizing the nature of the differentiating elec-
tron added to create a given atom from its predecessor, Bohr
concluded that all three transition series (though, as we saw,
he never used this term) began with the Sc group and ended
at the Ni group. The spectra of Cu and Zn were clearly analo-
gous to those of the alkali metals and alkaline earth metals,
respectively (22). By the 1950s, most inorganic textbooks had
adopted Bohr’s classification criterion rather than that of Bury,
even though they continued to use Bury’s terminology (18,
19). When stated in terms of the s, p, d, f orbital notation
later introduced by Hund, this criterion defines an outer-tran-
sition element as a simple substance containing atoms hav-
ing an incomplete (n

−

1)d subshell (23, 24). Given the

[NG](n

−

1)d

and the [NG](n

−

1)d

configurations

of the Cu and Zn groups (where [NG] stands for a noble gas
or pseudo-noble gas core), neither qualify by this definition
as transition elements (Table 1, line 4).

It is of interest to note the conflict between the chemi-

cal and spectroscopic definitions of a transition element. On
the basis of the chemical criteria given by Lewis and Bury,
the elements of the Cu group qualify as transition metals,
whereas as on the basis of the spectroscopic definition given
by Bohr and Hund, they are main-block elements. On the
basis of either definition, the metals of the Zn group are un-
ambiguously identified as main-block elements.

What Is a d-Block Element?

Given that most chemists now use the terms “transition

element” and “d-block element” interchangeably, the reader
might well wonder why this question deserves a separate sec-
tion (25). However, there are some authors who claim that
these two terms are not synonymous. Thus W. C. Fernelius,
in a 1986 review of labelling problems and the periodic table,
insisted that, although the members of the Zn group were
not transition elements, they were nevertheless d-block ele-
ments. He further stated that this latter concept had a “defi-
nite and consistent meaning”, though he failed to indicate
just what that meaning was (26). A survey of the three dozen
or so inorganic texts in my office also failed to turn up an
unambiguous definition.

One possibility is to define a d-block element as an ele-

ment whose differentiating electron occupies an (n

−

1)d or-

bital. A second possibility, following that given by Daintith,
is to define it as any element having a [NG](n

−

1)d

con-

figuration, where x can vary from 1–10 and z can vary from
0–2 (27). The first definition would preclude the Zn group,
for which the differentiating electron occupies an ns orbital,
whereas the second definition would include the Zn group,
as well as those transition elements having irregular configu-
rations (e.g., Pd). But while this last definition does repro-
duce the d block as it is shown in Figure 1, it appears to have
been invented after the fact to rationalize the frequent inclu-
sion of the Zn group among the transition metals, and it is
highly questionable whether it has either chemical or spec-
troscopic significance.

Chemistry is based on a differentiation between valence

electrons and core electrons, and it is the former that are of

‘

“

–

Information • Textbooks • Media • Resources

JChemEd.chem.wisc.edu • V ol. 80 No. 8 August 2003 • Jour nal of Chemical Education

955

most importance in defining the chemistry of a given ele-
ment and in determining its assignment to a given group of
the periodic table. As we have seen, spectroscopic character-
ization of the valence electrons in these elements shows that,
in the case of the neutral atoms, the significant break occurs
between the Ni group and the Cu group, though this fails to
coincide with a significant break in chemical properties. I,
on the other hand, would argue that the chemically significant
indicator is not whether an atom has a filled versus a partly
filled (n

−

1)d subshell, but whether the (n

−

1)d electrons from

this subshell can function as valence versus core electrons. In other
words, we can define a d-block element as any element that
uses either (n

−

1)d electrons or empty (n

−

1)d orbitals in

its bonding. This definition automatically precludes elements
in which the (n

−

1)d electrons have ceased to function as

valence electrons and have instead become a part of the
atomic core.

Bury and Lewis considered the (n

−

1)d subshell to al-

ways be a part of the atomic core. It merely functioned as a
reservoir of electrons that could be easily transferred to the
outermost or true valence shell. By contrast, the above defi-
nition considers the (n

−

1)d subshell to be a part of the va-

lence shell as long as there is evidence for the involvement of
(n

−

1)d electrons or orbitals in chemical bonding.

In lower oxidation-state species, that portion of the

−

1)d valence-electron density not involved in bond for-

mation is stored “internally” and consequently is generally
not stereoactive. As noted earlier, this is in sharp contrast to
the main-block elements in which that portion of the valence-
electron density not involved in bond formation is generally
stored—with the exception of the so-called inert pair effect—
in the form of stereoactive lone pairs. Hence, when species
containing a main-block atom act as electron-density donors
or Lewis bases, the structure of the isolated base undergoes
only minor perturbations upon neutralization. For example,

in the case of neutralization of ammonia by a proton, the
arrangement of electrons pairs around the N atom remains
essentially tetrahedral:

:NH

→

(1)

On the other hand, when a transition-metal complex acts as
a Lewis base, it undergoes substantial structural rearrange-
ment as a result of the stereochemical activation of part of
the original nonbonding valence-electron density (28). Thus,
in the case of the neutralization of the tetracarbonylcobalt
anion by a proton, the arrangement of the electron density
around the Co atom changes from tetrahedral to trigonal
bipyramidal:

Co(CO)

−

→

HCo(CO)

(2)

]

[

]

[

]

−

)

[

]

[

]

[

]

(

–

]

[

]

[

]

[

–

)

(

)

(

–

Information • Textbooks • Media • Resources

956

Journal of Chemical Education • V ol. 80 No. 8 August 2003 • JChemEd.chem.wisc.edu

Where Does the d-Block End?

All of the elements from the Sc group through the Cu

group exhibit one or more oxidation states in which the ele-
ment in question meets the above conditions and thus qualify
as d-block elements. The members of the Zn group, how-
ever, do not. This is best seen by comparing the properties
of the Cu group with those of the Zn group (Table 3). With
the exception of their M(I) oxidation states, all of the mem-
bers of the Cu group form high-oxidation-state compounds
that require use of one or more (n

−

1)d electrons, thus indi-

cating that, in spite of the filled (n

−

1)d

subshells in the

configurations of their neutral atoms, this subshell can still
serve as a source of valence-electron density (29). Even in
the case of the M(I) oxidation state, there is now evidence
that the filled (n

−

1)d

subshells are involved in weak metal-

metal bonding. This appears to be due to a mutual polariza-
tion of the filled subshells and is most pronounced in species
containing low-lying excited states capable of mixing with
the ground state. Evidence for these homonuclear d

–d

in-

teractions is particularly strong for the members of the Cu
group and has been reviewed by Jansen (30) and more re-
cently by Pyykö (31).

In sharp contrast, the members of the Zn group con-

tribute only two valence electrons to the bonding in their
known compounds. In no case is there any convincing evi-
dence that either (n

−

1)d electrons or empty (n

−

1)d or-

bitals are involved in their bonding interactions, thus
suggesting that, in contrast to the Cu group, the (n

−

1)d

subshell is now part of the atomic core, just as it is for the

case of those p-block elements that follow the Zn group in
the periodic table (32, 56). This inference has also been con-
firmed by theoretical calculations. At the end of their mo-
lecular orbital study of the changing role of the (n

−

1)d

orbitals in ZnS, FeS, and CrS, Hinchliffe and Dobson con-
cluded that (33):

Whereas the bonding in CrS involves mainly the partly
filled 3d orbitals, and in FeS a combination of 3d and
4s, in ZnS the “valence orbitals” on Zn are 4s and the
3d electrons behave as “core” electrons.

Where Does the Zn Group Fit into the Main Block?

As we have seen, from a spectroscopic point of view, the

members of the Zn group are analogous to the alkaline earth
metals or group II elements. From a chemical point of view,
Zn and Cd most resemble Be and Mg, not only in terms of
their atomic radii, ionic radii, and electronegativities (Table
4), but also in terms of the structures of their binary com-
pounds and in their ability to form complex ions with a wide
variety of oxygen and nitrogen donor ligands (including com-
plex hydrates and amines). Indeed, prior to the introduction
of electronic periodic tables, the similarity between Be and
Mg and Zn and Cd was often considered to be greater than
the similarity between Be and Mg and the rest of the alka-
line earth metals (Ca–Ra). Many inorganic texts written be-
fore the Second World War placed their discussion of the
chemistry of Be and Mg in the chapter dealing with the Zn
subgroup rather than in the chapter dealing with the Ca sub-
group, and the same is true of many older periodic tables,
including those originally proposed by Mendeleev (34, 35).
Even as late as 1950, N. V. Sidgwick, in his classic two-vol-
ume survey of The Chemical Elements and Their Compounds,
felt that it was necessary to justify his departure from this
scheme in the case of Mg (36):

The gap between magnesium and the succeeding ele-
ments is sufficient to make it desirable to treat the mag-
nesium compounds separately, but in the general
discussion we may include it along with the alkaline
earths metals proper, the elements from calcium to ra-
dium.... There has been much argument as to whether
magnesium has the properties of a member of group IIA
(alkaline earth metals) or IIB (zinc, cadmium, mercury);
in fact of course it has resemblances to both. It is found
in nature rather with the A elements than with the B. In
its chemistry it shows analogies to both subgroups; in its
power of complex formation it stands between the two,
and seems to come nearer to B than A, but this is a natural
result of its small size....

As suggested by Sidgwick’s final comment, these simi-

larities are easily understood as a consequence of having in-
serted the d block between the Ca and Zn subgroups. This
results in an enhancement of the effective nuclear charges (Z*)
for Zn and Cd over those of Ca–Ra and leads to values for
their radii and electronegativities that are much closer to the
values found for Be and Mg than are those of the Ca sub-
group. In the case of Hg, the combination of both the d-
block and f-block insertions, as well as significant relativistic
effects, results in properties that are virtually unique, though

Figure 4. A plot of group electronegativity trends for the main block
elements (modified Martynov–Batsanov scale) showing the bifurca-
tion at Mg. If one follows the Ca–Ra branch at Mg, the trends par-
allel those of the alkali metals in group I. If, instead, one follows
the Zn–Hg branch at Mg, the trends parallel those shown in groups
III–VIII. This change in pattern is due to the insertion of the d and f
blocks and the resulting increase in

Z* accounts for the enhanced

electronegativities of the Zn–Hg branch over those of the Ca–Ra
branch and for the similarity between their chemistry and that of
Mg and Be.

Information • Textbooks • Media • Resources

JChemEd.chem.wisc.edu • V ol. 80 No. 8 August 2003 • Jour nal of Chemical Education

957

this element still retains the formal similarity of having only
two outer ns-type valence electrons (37).

In general, if one plots a given property for the mem-

bers of group II, the group trend will parallel that shown by
the alkali metals in group I if, after Mg, one follows the Ca–
Ra subgroup or branch, but will parallel the pattern shown
by groups III–VIII if one follows the Zn–Hg subgroup or
branch. This is illustrated in Figure 4 for group trends in elec-
tronegativity. In summary, both the chemical and spectro-
scopic evidence clearly place the Zn group in group II of the
main-block elements and lead to the irresistible conclusion
that, following Be and Mg, there is a fundamental bifurca-
tion of this group into a Ca–Ra branch and a Zn–Hg branch.

Representing Affiliation by Position

Given the above conclusion, how can we indicate this

bifurcation of group II on a periodic table? As long as chem-
ists preferred to use the short 8-column form of the table,
this was not a problem, since, beginning with period 4, all
eight groups automatically bifurcated into an A and B branch
or subgroup. This is best shown in the version of the short
table (Figure 5) proposed by Venable in 1895 and again by
Sears in 1924 (38, 39). The pairs of elements found in peri-
ods 2 and 3 and located at the head of each group were known
as the “typical elements”, and many short tables also made
some attempt to indicate whether these pairs most resembled
their corresponding A or B subgroups, usually by preferen-
tially aligning them over one branch or the other (40).

It was only with the increasing preference for the longer

18- and 32-column tables that the placement of Zn group
or, more accurately, the placement of Be and Mg became a
problem. These expanded tables destroyed the alignment of
most of the subgroups by spatially separating the transition
elements from the main-block elements. While this conse-
quence is considered a great improvement, it simultaneously
separated the main-block Ca and Zn subgroups as well and
raised the question of whether Be and Mg should be aligned
with the Ca–Ra branch or with the Zn–Hg branch.

This problem is well illustrated by a debate that occurred

between Alfred Werner and Richard Abegg. In April of 1905,
Werner proposed a 32-column periodic table in which he
followed Mendeleev in placing Be and Mg above the Zn sub-
group rather than above the Ca subgroup (Figure 6; ref 41,
42). Abegg, who was a strong advocate of the short table,
wrote a rejoinder in which, among other things, he objected
to this placement (43, 44). His only contrary argument was
that, if he plotted the change in properties down the Be–Ra
alternative, he obtained regular trends, whereas, if he plot-
ted them down the Be–Hg alternative, he obtained irregu-
larities at both Zn and Hg. However, as we have already seen,
this difference is a consequence of the d- and f-block inser-
tions and the fact that the group trends for the Be–Ra choice
parallel those of group I whereas those for the Be–Hg choice
parallel the trends for groups III–VIII.

This exact same debate was repeated nearly 20 years later

when Fritz Paneth published a paper in 1923 on new ways
of formulating the periodic table (45). In his proposed 18-

Figure 6. An updated ver-
sion of Werner’s 32-column
block-table of 1905 show-
ing his placement of Be and
Mg in the Zn group.

Figure 5. An updated
Venable–Sears modification
of the 8-column short table
given in Figure 2, empha-
sizing the bifurcation of
each group into A and B
branches or subgroups.

Information • Textbooks • Media • Resources

958

Journal of Chemical Education • V ol. 80 No. 8 August 2003 • JChemEd.chem.wisc.edu

column table (Figure 3), he placed Be and Mg above the Ca
subgroup rather than above the Zn subgroup. This elicited a
response the next year from Paul Pfeiffer, who presented de-
tailed chemical arguments, similar to those given above, for
why these two elements should be aligned with the Zn sub-
group instead (46). Paneth’s only substantive counter-argu-
ment was that the members of Zn subgroup had a filled (n

−

1)d

subshell in their configurations that was absent in the

case of both Be and Mg and in the members of the Ca sub-
group (47). But again, as we have already seen, these d elec-
trons function as core electrons rather than as valence
electrons. Just as the presence of filled (n

−

1)d

subshells in

the cores of the elements following the Zn group in no way
disqualifies them from being classified as p-block elements,
so their presence in the cores of the Zn-group elements in
no way disqualifies them from being classified as s-block ele-
ments.

There is virtually no chance that the short 8-column

form of the periodic table will ever regain its former popu-
larity. However, there is at least one form of the 32-column
table that allows one to maintain the spatial affiliations of
the old subgroups while simultaneously separating the tran-
sition metals from both the main-block elements and from

the lanthanoids and actinoids. This is the famous step-pyra-
mid table first proposed by Thomas Bailey in 1882 and again
by Julius Thomsen in 1895 (48, 49). This table saw exten-
sive textbook usage in the period 1925–1945, largely as a
result of its popularization by Niels Bohr (22). Group affili-
ation in this table is indicated either by vertical alignment or
by means of diagonal tie lines, thus allowing one to indicate
simultaneously multiple relationships (50). As can be seen
in the updated version given in Figure 7, Mg is simultaneously
connected to both the Ca and Zn subgroups by means of
solid diagonal lines, thus indicating, in keeping with the facts
summarized earlier by Sidgwick, equal rather than preferen-
tial affiliation, and thus perfectly resolving our conflict.

In my opinion, the step-pyramid form of the periodic

table is indisputably superior to the common 18-column
block table used in virtually all of our textbooks, especially
in the case of advanced inorganic courses in which these sub-
tler chemical relationships become important. In the case of
introductory classes, however, I have encountered a certain
resistance on the part of students to this form of the table.
Much of this is, of course, due to the fact that they have al-
ready become familiar with the 18-column block table in their
high school chemistry courses and do not want to switch

Figure 7. A modernized step-pyramid table showing the bifurcation of Group II following Mg. Numbers to the left of the rows represent the
periods. The “e + v” numbers indicate the total number of valence electrons (e) and valence vacancies (v) per electronic block.

Information • Textbooks • Media • Resources

JChemEd.chem.wisc.edu • V ol. 80 No. 8 August 2003 • Jour nal of Chemical Education

959

gears. But they also seem to be bothered by the use of diago-
nal as well as vertical alignments and by the increasing sepa-
ration between the s-block and p-block elements as one moves
from the top to the bottom of the table. Since the descrip-
tive chemistry of the introductory course is largely restricted
to these two blocks of elements, the students tend to view
the d- and f-block insertions as unnecessary distractions.

An elegant solution to these problems was proposed by

R. T. Sanderson more than 30 years ago (51, 52). Just as we
shorten the full 32-column periodic table into an 18-column
table by pulling out the f-block elements and attaching them
as an appendix at the bottom of the table, so Sanderson did
the same for the d-block, thus generating what might prop-
erly be called a “double-appendix” table. A slightly revised
form of this table created from the step-pyramid table in Fig-
ure 7 is shown in Figure 8. As can be seen, this table cor-
rectly shows the termination of the d block at the Cu group,
the termination of the f block at the Yb group rather than at
the Lu group (53), and the placement of both the Ca and
Zn subgroups together in group II of the main block.

Even more important from a pedagogical standpoint,

Sanderson’s table has decomposed the periodic table into three
rectangular subtables—one for the main-block or sp-block

elements, one for the transition or d-block elements, and one
for the lanthanoid–actinoid or f-block elements. Each
subtable has its own internal consistency and can be selec-
tively emphasized at different stages of the curriculum—the
main-block subtable in introductory courses, the transition-
block subtable in intermediate inorganic courses, and the f-
block subtable in advanced inorganic courses.

Despite its extraordinary advantages, Sanderson’s double

appendix table has seen virtually no use beyond his own writ-
ings (54). It is unclear whether this is due to resistance on
the part of authors and publishers, who fear that any depar-
ture from the norm will diminish the sale of their textbooks,
or to the fact that the use of the periodic table to correlate
the facts of descriptive chemistry is so superficial in most text-
books that the very real limitations of the 18-column block
table never become apparent.

Representing Affiliation with Labels

Affiliation can, of course, be indicated by group labels

as well as by spatial alignment. In the original European AB
labelling system (Figure 3, third row of labels), the Zn sub-
group was labeled IIB and the Ca subgroup was labeled IIA.
The A and B modifiers were intended to indicate that the
Ca subgroup belonged to the first valence cycle of the long
periods and the Zn subgroup to the second valence cycle,
respectively. However, to students and chemists who are un-
familiar with the logic behind this scheme, it appears that
the Zn group is being incorrectly classified with the succeed-
ing p-block elements, all of which also carry the B modifier.
In the American ABA system (Figure 3, second row of la-
bels), the calcium and zinc subgroups are again labeled IIA
and IIB, respectively, but the modifiers now refer to main-
block (A) versus transition-block (B) elements instead and
thus incorrectly classify the Zn group with the transition
metals. Hence neither of the traditional modifier systems is
satisfactory.

A much better scheme was suggested by Sanderson and

is shown in Figure 8. This scheme uses an MT modifier sys-
tem to indicate whether a given group of elements belongs
to the main block or major groups (M) or to the transition
block (T). In Sanderson’s system the Ca and Zn subgroups
are labeled M2 and M2', respectively, thus clearly indicating
the fundamental bifurcation of group 2. Examination of the
textbook by Day and Selbin (who use R for representative
element and label the Ca and Zn subgroups R2 and R2' in-
stead) shows that modifier systems of this type can be used
on the standard 18-column block table (Figure 3, top row of
labels) to correctly indicate the placement of the Zn group
even in the absence of the spatial alignment provided by the
alternative tables discussed in the previous section (55).

Unhappily, the recent decision of IUPAC to substitute

enumerator labels in place of descriptive labels would appear
to rule out Sanderson’s solution. Their suggested 1–18 labels
(Figure 3, fourth row of labels) provide only “finger count”
information about the number of columns in the periodic
table but tell us nothing about the chemistry or logic of the
classification scheme. Although this decision may make com-
puter indexing easier, it also tends to disguise some funda-
mental problems, such as the Be and Mg placement discussed
above, and thus diminishes our understanding of chemistry.

Figure 8. An updated version of Sanderson’s double-appendix table
showing the proper location of the Zn group and the proper termi-
nation of both the d block and the f block. Sanderson did not pro-
vide group labels for the f-block elements

Information • Textbooks • Media • Resources

960

Journal of Chemical Education • V ol. 80 No. 8 August 2003 • JChemEd.chem.wisc.edu

Summary

Forcing the Zn group into the transition or d block, be-

cause the resulting periodic tables are thought to be more
symmetric and easier for students to memorize, is in many
ways the 20th century equivalent of the 19th century’s deci-
sion to force the Cu group into group I along with the alkali
metals. We all realize that, for the sake of our students, we
often need to simplify the material we teach, but simplifying
a theory or model by leaving out details is very different from
falsifying the facts (however idiosyncratic) of chemical be-
havior.

Literature Cited

1. Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M.

Advanced Inorganic Chemistry, 6th ed.; Wiley: New York,
1999; See the table of contents and Chapter 15.

2. Holleman–Wiberg Inorganic Chemistry, 34th ed.; Wiberg, N.,

Ed.; Academic Press: New York, 2001; Chapters 19, 23.

3. Massey, A. G. Main Group Chemistry, 2nd ed.; Wiley: New

York, 2000; Chapters 2, 6.

4. Jones, C. D- and F-Block Chemistry; Royal Society of Chem-

istry: Cambridge, England, 2001.

5. Examples of advanced monographs that include the Zn group

among the transition metals include: Lukehart, C. M. Fun-
damental Transition Metal Organometallic Chemistry; Brooks/
Cole: Monterey, CA, 1985. Atwood, J. D. Inorganic and Or-
ganometallic Reaction Mechanisms; Brooks/Cole: Monterey, CA,
1985.

6. Examples of advanced monographs that exclude the Zn group

from the transition metals include: King, R. B. Transition-
Metal Organometallic Chemistry: An Introduction; Academic
Press: New York, 1969. Nicholls, D. Complexes and First-Row
Transition Elements; Macmillan: London, 1974.

7. For a typical example from a general chemistry text, see Atkins,

P. W. General Chemistry; Scientific American Books: New York,
1989, p 44, Figure 2.3. This incorrect classification of the
members of the Zn group as d-block elements is also adopted
within the text, see p 775.

8. For a typical example from an introductory inorganic text, see

Rayner-Canham, G. Descriptive Inorganic Chemistry; Freeman:
New York, 1996, p 20, Figure 2.3. This classification of the
members of the Zn group as d-block elements contradicts their
later treatment within the text, where they are correctly de-
scribed as main-block elements, see p 454.

9. Caven, R. M.; Lander, G. D. Systematic Inorganic Chemistry

From the Standpoint of the Periodic Law; Blackie & Son: Lon-
don, 1907; pp 53–54.

10. Mellor, J. W. Modern Inorganic Chemistry; Longmans: Lon-

don, 1927; p 993.

11. Partington, J. R. A Text-Book of Inorganic Chemistry;

Macmillan: London, 1939; pp 412–415, 955.

12. Bury, C. R. J. Am. Chem. Soc.

1921

, 43, 1602–1609.

13. Lewis, G. N. J. Am. Chem. Soc.

1916

, 38, 762–785.

14. Lewis, G. N. Valence and the Structure of Atoms and Molecules;

Chemical Catalog Co.: New York, 1923; pp 63–64.

15. Bohr, N. Nature

1921,

107, 104–107.

16. This statement is based on a survey of the papers and lectures

reprinted in Niels Bohr: Collected Works–The Periodic System

(1920–1923); Nielsen, J. R., Ed.; North-Holland Publishing:
Amsterdam, 1977; Vol. 4.

17. Compare, for example, the use of the term transition element

in the 1907 edition of ref 9 with that given in the postwar
edition of 1945. See, Caven, R. M.; Lander, G. D.; Crawford,
A. B. Systematic Inorganic Chemistry from the Standpoint of the
Periodic Law, 6th ed.; Blackie & Son: London, 1945; p 64.
This switch occurred much earlier in this country than in Eu-
rope.

18. Moeller, T. Inorganic Chemistry: An Advanced Textbook; Wiley:

New York, 1952; pp 102–103.

19. Gilreath, E. S. Fundamental Concepts of Inorganic Chemistry;

McGraw-Hill: New York, 1958; p 130.

20. Mendeleev, D. Ann. Chem. Pharm. (Supplement Band)

1871,

8, 133–229.

21. For a more detailed list of classical chemical and physical dif-

ferences between the main block and the transition elements,
see Yoe, J. H; Sarver, L. A. Organic Analytical Reagents; Wiley:
New York, 1941; pp 119–121.

22. Bohr, N. The Theory of Spectra and Atomic Constitution; Uni-

versity Press: Cambridge, 1922; p 70.

23. Hund, F. Linienspektren und periodisches System der Elemente;

Springer: Berlin, 1927; pp 52–54.

24. Day, M. C.; Selbin, J. Theoretical Inorganic Chemistry, 2nd ed.;

Reinhold: New York, 1969; p 100.

25. Eggleton, G. Transitional Elements. In Macmillan Encyclope-

dia of Chemistry; Lagowski, J. J., Ed.; Macmillan Reference:
New York, 1997; Vol. 4, pp 409–412. This recent article treats
the terms “transitional” and “d block” as synonyms and clas-
sifies the Zn group among the transition elements.

26. Fernelius, W. C. J. Chem. Educ.

1986,

63, 263–266. A simi-

lar distinction, without independent justification of the term
d block, appears in the recent textbook: Housecroft, K.;
Sharpe, A. G. Inorganic Chemistry; Pearson: Harlow, England,
2001; p 434.

27. Daintith, J. Dictionary of Chemistry; Barnes & Noble: New

York, 1981; p 63.

28. Shriver, D. F. Acc. Chem. Res.

1970,

3, 231–238.

29. For a recent update on the oxidation states of the Cu group,

see Müller, B. G. Angew. Chem., Int. Ed. Engl.

1987,

26, 1081–

1097.

30. Jansen, M. Angew. Chem., Int. Ed. Engl.

1987,

26, 1098–1110.

31. Pyykö, P. Chem. Rev.

1997,

97, 597–636.

32. Neither Jansen or Pyykö give detailed evidence for d

–d

interactions in the compounds of the elements of the Zn
group. The strongest evidence that Jansen could offer was to
suggest that the distortions of the metallic structures of these
elements from ideal hexagonal and cubic closest packing might
be due to this effect. However, the members of the Cu group
show no such distortions in their metallic structures, even
though they are well documented for their compounds. In any
case, we would expect such polarizations to decrease in im-
portance, if not totally disappear, as the (n

−

1)d

subshell

becomes more corelike in its behavior. See also
Balasubramanian, K. Relativistic Effects in Chemistry; Wiley:
New York, 1997.

33. Hinchliffe A.; Dobson, J. C. Theoret. Chim. Acta

1975,

39,

211–216.

34. See, for example, Mellor, J. W. Modern Inorganic Chemistry;

Longmans: London, 1927; Chapter 8. Partington, J. R. A

Information • Textbooks • Media • Resources

JChemEd.chem.wisc.edu • V ol. 80 No. 8 August 2003 • Jour nal of Chemical Education

961

Text-Book of Inorganic Chemistry; Macmillan: London, 1939;
Chapter 42. Hopkins, B. S. Chemistry of the Rarer Elements;
Heath: Boston, 1923, Chapter 5.

35. See, for example, the tables given in Mendeleev, D. Ann. Chem.

Pharm. (Supplement Band),

1871,

8, 133–229. Werner, A.

Berichte

1905,

38, 914–921.

36. Sidgwick, N. V. The Chemical Elements and their Compounds;

Clarendon Press: Oxford, England, 1950; pp 219, 222.

37. Norrby, L. J. J. Chem. Educ.

1991,

68, 110–111.

38. Venable, F. P. J. Am. Chem. Soc.

1895,

17, 75–84.

39. Sears, G. W. J. Chem. Educ.

1924,

1, 173–177.

40. For example tables, see Sears, G. W. J. Chem. Educ.

1924,

173–177. Letts, E. A. Some Fundamental Problems in Chemis-
try Old and New; Constable: London, 1914; p 54.

41. Werner, A. Berichte

1905,

38, 914–921.

42. Werner, A. Berichte

1905,

38, 2022–2027.

43. Abegg, R. Berichte

1905,

38, 1386–1388.

44. Abegg, R. Berichte

1905,

38, 2330–2334.

45. Paneth, F. Angew. Chem.

1923,

36, 407–410.

46. Pfeiffer, P. Angew. Chem.

1924,

37, 41.

47. Paneth, F. Angew. Chem.

1924,

37, 421–422.

48. Bailey, T. Phil. Mag.

1882,

13 (5), 26–37.

49. Thomsen, J. Z. Anorg. Chem.

1895,

9, 190–193.

50. Jensen, W. B. Educ. Chem.

1996,

32, 122–123.

51. Sanderson, R. T. J. Chem. Educ.

1964,

41, 187–189.

52. Sanderson, R. T. Inorganic Chemistry; Reinhold: New York,

1967; pp 14–16.

53. Jensen, W. B. J. Chem. Educ.

1982,

59, 634–636.

54. For an exception, see Pode, J. The Periodic Table; Halsted: New

York, 1970; p 21.

55. See table in Day M. C.; Selbin, J. Theoretical Inorganic Chem-

istry, 2nd ed.; Reinhold: New York, 1969; p 84.

56. There has also been an unsubstantiated claim that it is pos-

sible to electrochemically generate a Hg(cyclam)

cation in

acetonitrile at

᎑

⬚

C with a lifetime of about 5 seconds. For

details see Cotton, F. A.; Wilkinson, G.; Murillo, C. A.;
Bochmann, M. Advanced Inorganic Chemistry, 6th ed.; Wiley:
New York, 1999.

57. Noyes, A. A.; Bray, W. C. A System of Qualitative Analysis for

the Rare Elements; Macmillan: New York, 1927; p 8.