Atoms


Proust's Law

Lavoisier's successes stimulated chemists to search out and explore other areas in which accurate measurements might illuminate the study of chemical reactions. The acids comprised one such area.

Acids form a natural group sharing a number of properties. They tend to be chemically active, reacting with metals such as zinc, tin, or iron, dissolving them and producing hydrogen. They taste sour (if dilute enough or weak enough to be tasted with impunity), cause certain dyes to change colors in certain ways, and so on.

Opposed to the acids is another group of substances called bases. (Strong bases are termed alkalis). These are also chemically active, taste bitter, change dye colors in a fashion opposite to that induced by acids, and so on. In particular, solutions of acids will neutralize solutions of bases. In other words, if acids and bases are mixed in proper proportions, then the mixture will show the property of neither acids nor bases. The mixture will be, instead a solution of a salt, which, in general, is a much milder chemical than either an acid or a base. Thus, a solution of the strong and caustic acid, hydrochloric acid, if mixed with the proper amount of the strong and caustic alkali, sodium hydroxide, will become a solution of sodium chloride, ordinary table salt.

The German chemist Jeremias Benjamin Richter (1762-1807) turned his attention to these neutralization reactions, and measured the exact amounts of different acids that were required to neutralize a given quantity of a particular base, and vice versa. By careful measurements he found that fixed and definite amounts were required. There wasn't the leeway that a cook might count on in the kitchen, where a bit more or less of some ingredient is not terribly important. Instead, there was such a thing as an equivalent weight; a fixed weight of one chemical reacted with a fixed weight of another chemical. Richter published his work in 1792.

Two French chemists were then engaged in strenuous battle over whether this sort of definiteness existed not only in acid-base neutralization but throughout chemistry. To put it fundamentally, if a particular compound were made up of two elements (or three or four), were those two elements (or three or four) always present in this compound in the same, fixed proportions? Or would these proportions vary, depending on the exact method of preparing the compound? Berthollet, one of those who collaborated with Lavoisier in establishing modern chemical terminology, thought the latter. According to Berthollet's view, if a compound consisted of element x and y, then it would contain a more then average quantity of x, if it were prepared while using x in large excess.

Opposed to Berthollet's view was the opinion of Joseph Louis Proust (1754-1826), who did his work in Spain, safe (for a time) from the upheavals of the French Revolution. Using painstakingly careful analysis, Proust showed, in 1799, that copper carbonate, for instance, contained definite proportions by weight of copper, carbon, and oxygen, no matter how it was prepared in the laboratory or how it was isolated from natural sources. The preparation was always 5.3 parts of copper to 4 of oxygen to 1 of carbon.

Proust went on to show that a similar situation was true for a number of other compounds, and formulated the generalization that all compounds contained elements in certain definite proportions and in no other combinations, regardless of the conditions under which they were produces. This is called the law of definite proportion or, sometimes, Proust's Law. (Proust also showed that Berthollet, in presenting evidence that certain compounds varied in composition according to the method of preparation, was misled through inaccurate analysis and through the use of products he had insufficiently purified).

During the first few years of the nineteenth century, it became quite clear that Proust was right. Other chemists verified the law of definite proportions, and it became a cornerstone of chemistry. (It is true that some substances can vary, within limits, in their elemental constitution. These are special cases. The simple compounds which engaged the attention of the chemists of 1800 held firmly to the law of definite proportions).

From the moment Proust's law was announced, serious thoughts concerning it were forced into the chemical view.

After all, why should the law of definite proportions hold true? Why should a certain compound be made up always of 4 parts x and 1 part y, let us say, and never of 4.1 parts x or 3.9 parts x to 1 part y. If matter were continuous, this would be hard to understand. Why could not elements be mixed in slightly varying proportions?

But what if matter was atomistic in nature? Suppose a compound was formed when one atom of x joined with one atom of y and not otherwise. (Such a combination of atoms eventually came to be called a molecule, from a Latin word meaning "a small mass"). Suppose, next, that each atom of x happened to weigh four times as much as each atom of y. The compound would then have to consist of exactly 4 parts of x to 1 part of y.

In order to vary those proportions, an atom of y would have to be united with slightly more or slightly less than one atom of x. Since an atom, ever since the time of Democritus, had been viewed as being an indivisible portion of matter, it was unreasonable to expect that a small piece might be chipped off an atom, or that a sliver of a second atom might be added to it.

In other words, if matter consisted of atoms, then the law of definite proportions followed as a natural consequence. Furthermore, from the fact that the law of definite proportions was an observed fact, one could deduce that atoms were indeed indivisible objects.

Dalton's Theory

An English chemist, John Dalton (1766-1844), went through this chain of reasoning. In this, he was greatly aided by a discovery he made. Two elements, he found, might, after all, combine in more than one set of proportions, but in so doing they exhibited a wide variation of combining proportions and different compound was formed for each variation.

As a simple example, consider the elements carbon and oxygen. Measurement shows that 3 parts of carbon (by weight) will combine with 8 parts of oxygen to form carbon dioxide. However, 3 parts of carbon and 4 parts of oxygen make up carbon monoxide. In such a case, the differing quantities of oxygen that combine with a fixed amount of carbon are found to be related in the form of small whole numbers. The 8 parts present in carbon dioxide is exactly twice that of the 4 parts present in carbon monoxide.

This is the law of multiple proportions. Dalton, after observing its existence in a number of reactions, advanced it in 1803.

The law of multiple proportions fits in neatly with atomistic notions. Suppose, for instance, that atoms of oxygen are uniformly 1-1/3 times as heavy as atoms of carbon. If carbon monoxide is formed through the combination of one atom of carbon with one atom of oxygen, the compound must consist of 3 parts by weight of carbon to 4 parts of oxygen.

Then, if carbon dioxide is formed of one atom of carbon and two atoms of oxygen, the proportion must naturally consist of 3 parts of carbon to 8 of oxygen.

The relationship in simple multiples would reflect the existence of compounds varying in makeup by whole atoms. Surely, if matter did indeed consist of tiny, indivisible atoms, these would be just the variations in makeup you would expect to find, and the law of multiple proportions makes sense.

When Dalton put forward his new version of the atomic theory based on the laws of definite proportions and of multiple proportion, in 1803, he acknowledged the debt to Democritus by keeping the term "atom" for the small particles that made up matter.

In 1808, he published A New System of Chemical Philosophy, in which his atomic theory was discussed in greater detail. In that year, too, his law of multiple proportions was verified by the investigations of another English chemist, William Hyde Wollaston (1766-1828). Wollaston lent his influential weight to the atomic theory in consequence, and Dalton's view in due course won general acceptance.

The atomic theory, by the way, was a death blow (if any were needed) to belief in the possibility of transmutation of alchemical terms. All evidence seemed to point to the possibility that the different metals each consisted of a separate type of atom. Since atoms were taken generally to be indivisible and unchangeable, one could not expect to change a lead atom to a gold atom in any circumstances. Lead, therefore, could not be transmuted to gold. (A century after Dalton's time this view had to be modified. One atom could, after all, be changed to another. The methods used to achieve this, however, were such as no alchemist ever imagined or could have performed.)

Dalton's atoms were, of course, far too small to be seen even under a microscope; direct observation was out of the question. Indirect measurements, however, could yield information as to their relative weights.

For instance, 1 part (by weight) of hydrogen combined with 8 parts of oxygen to form water. If one assumed that a molecule of water consisted of one atom of hydrogen and one atom of oxygen, then it would follow that the oxygen atom was eight times as heavy as the hydrogen atom. If it was decided to set the weight of the hydrogen atom arbitrarily equal to 1, then the weight of the oxygen atom on that scale would be 8.

Again, if 1 part of hydrogen combines with 5 parts of nitrogen in forming ammonia, and it is assumed that the ammonia molecule is made up of one atom of hydrogen and on of nitrogen, it would follow that the nitrogen atom would have a weight of 5.

Reasoning after this fashion, Dalton set up the first table of atomic weights. This table, although perhaps his most important single contribution, proved to be quite wrong in many entries. The chief flaw lay in Dalton's insistence that in general molecules were formed by the pairing of a single atom of one element with a single atom of another. He varied from this position only when absolutely necessary.

Evidence piled up, however, that indicated such a one-to-one combination was not necessarily the rule at all. The disagreement showed up in connection with water, in particular, even before Dalton had advanced his atomic theory.

Here, for the first time, the force of electricity invades the world of chemistry.

Knowledge of electricity dates back to the ancient Greeks, who found that when amber is rubbed, it gains the power to attract light objects.

Centuries later, the English physicist William Gilbert (1540-1603) was able to show that it was not amber alone that acted so, but that a number of other substances as well gained an attracting power when rubbed. About 1600, he suggested that substances of this sort be called "electrics", from the Greek word for amber.

As a result, a substance that gains such a power, through rubbing or otherwise, is said to carry an electric charge, or to contain electricity.

The French chemist Charles Francois de Cisternay du Fay (1698-1739) discovered, in 1733, that there were two kinds of electric charge: one that could be put on glass ("vitreous electricity") and one that could be put on amber ("resinous electricity"). A substance carrying one kind of charge attracted another substance carrying the other, but two substances bearing the same kind of charge repelled each other.

Benjamin Franklin (1706-1790), who was the first great American scientist as well as a great statesman and diplomat, suggested, in the 1740's, that there was a single electrical fluid. When a substance contained a greater than normal quantity of electric fluid, it possessed one kind of electric charge; when it contained a less than normal quantity, it possessed the other kind.

Franklin guessed it was the glass that contained the greater than normal quantity of electric fluid, so he said it carried a positive charge. The resin, he said, carried a negative charge. Franklin's terms have been used ever since, although the usage leads to a concept of current flow opposite to what now is known to be the fact.

The Italian physicist Alessandro Volta (1745-1827) introduced something new. He found, in 1800, that two metals (separated by solutions capable of conducting an electric charge) could be so arranged that new charge was created as fast as the old charge was carried off along a conducting wire. He had invented the first electric battery and produced an electric current.

Such an electric current is maintained by the chemical reaction involving the two metals and the solution between. Volta's work gave the first clear indication that chemical reactions had something to do with electricity, a suggestion that was not to be developed completely for another century. If a chemical reaction could produce an electric current, it did not seem to be too farfetched to suppose that an electric current could reverse matters and produce a chemical reaction.

Indeed, within six weeks of Volta's first description of his work, two English chemists, William Nicholson (1753-1815) and Anthony Carlisle (1768-1840), demonstrated the reverse action. They ran an electric current through water and found bubbles of gas began to appear at the electricity-conducting strips of metal which they had inserted in the water. The gas appearing at one strip was hydrogen and that appearing at the other was oxygen.

In effect, Nicholson and Carlisle had decomposed water into hydrogen and oxygen, such decomposition by an electric current being called electrolysis. They had achieved the reverse of Cavendish's experiment, in which hydrogen and oxygen had been combined to form water.

When the hydrogen and oxygen were trapped in separate vessels as they bubbled off, it turned out that just twice as large a volume of hydrogen was formed as of oxygen. The hydrogen was the lighter in weight, to be sure, but the larger volume indicate that there might be more atoms of hydrogen than of oxygen in the water molecule.

Since there was just twice as large a volume of hydrogen produced as of oxygen, there was at least a certain reasonableness in supposing that each molecule of water contained two atoms of hydrogen and one of oxygen, rather than on of each, as Dalton proposed.

Even if this were so, it remained true that 1 part of hydrogen (by weight) was combined with 8 parts of oxygen. It followed, then, that one oxygen atom was eight times as heavy as two hydrogen atoms taken together, and, therefore, sixteen times as heavy as a single hydrogen atom. If the weight of hydrogen is set at 1, then, the atomic weight of oxygen must be 16, not 8.

Avogadro's Hypothesis

The findings of Nicholson and Carlisle were strengthened by the work of a French chemist, Joseph Louis Gay-Lussac (1778-1850), who reversed matters. He discovered that 2 volumes of hydrogen combined with 1 volume of oxygen to form water. He went on to find, in fact, that when gases combine to form compounds, they always did so in small whole number ratios. Gay-Lussac announced this law of combining volumes in 1808.

From the whole numbers ratios in the formation of water from hydrogen and oxygen, it again seemed reasonable to suppose that the water molecule was composed of two atoms of hydrogen and one of oxygen. It could also be argued from similar lines of evidence that the ammonia molecule did not consist of a combination of one nitrogen atom and one hydrogen atom, but of one nitrogen atom and three hydrogen atoms. From that evidence one could conclude that the atomic weight of nitrogen was not nearly 5, but was 14.

Consider hydrogen and chlorine next. These are gases which combine to form a third gas, hydrogen chloride. One volume of hydrogen combines with one volume of chlorine, and it seems reasonable to suppose that the hydrogen chloride molecule is made up of one hydrogen atom combined with one chlorine atom.

Suppose, now, that the hydrogen gas consists of single hydrogen atoms, spaced widely apart, and the chlorine gas consists of single chlorine atoms, spaced equally widely apart. These atoms pair up to form hydrogen chloride molecules, also spaced equally widely apart.

We begin, let us say, with 100 atoms of hydrogen and 100 atoms of chlorine, giving 200 widely spaced particles all told. The atoms pair up to form 100 molecules of hydrogen chloride. The 200 widely spaced particles (atoms) become only 100 widely spaced particles (molecules). If the spacing is equal throughout, we should find that 1 volume of hydrogen plus 1 volume of chlorine (2 volumes, altogether) should yield only 1 volume of hydrogen chloride. This, however, is not so.

By actual measurement, 1 volume of hydrogen combines with 1 volume of chlorine to form 2 volumes of hydrogen chloride. Since 2 volumes to start with remain 2 volumes to end with, there must be the same number of widely spaced particles before and after.

But suppose the hydrogen gas exists not as separate atoms but as hydrogen molecules, each made up of 2 atoms, and that chlorine consists of chlorine molecules, each made up of 2 atoms. In that case, the 100 atoms of hydrogen exist as 50 widely spaced particles (molecules), and the 100 atoms of chlorine also exist as 50 widely spaced particles. In the two gases, together, there are 100 widely spaced particles altogether, half of them hydrogen-hydrogen and the other half chlorine-chlorine.

If the two gases combine, they rearrange themselves to form hydrogen-chlorine, the atomic combination making up the hydrogen chloride molecule. Since there are 100 atoms of hydrogen altogether and 100 atoms of chlorine, there are 100 molecules of hydrogen chloride (each containing one of each kind of atom).

Now we find that 50 molecules of hydrogen plus 50 molecules of chlorine combine to form 100 molecules of hydrogen chloride. This matches the actually observed 1 volume of hydrogen plus 1 volume of chlorine yielding 2 volumes of hydrogen chloride.

All this takes for granted the fact that the particles of different gases, whether composed of single atoms or of combinations of atoms, are indeed equally spaced apart. If so, then equal numbers of particles of a gas (at a given temperature) would take up equal volumes no matter what the gas is.

The first to point out the necessity of this assumption that, in gases, equal numbers of particles take up equal volumes, was the Italian chemist Amedeo Avogadro (1776-1856). The assumption, advanced in 1811, is therefore known as Avogadro's hypothesis.

If the hypothesis is kept firmly in mind, it is possible to distinguish clearly between hydrogen atoms and hydrogen molecules (a pair of atoms) and between the atoms and molecules of other gases, too. For half a century after Avogadro's time, however, his hypothesis lay neglected, and the distinction between atoms and molecules of the important gaseous elements was not clearly defined in the minds of most chemists. Considerable uncertainty as to the value of the atomic weights of some of the most important elements persisted.

Fortunately, there were other keys to correctness in atomic weights. In 1818, for instance, a French chemist, Pierre Louis Dulong (1785-1838), and a French physicist, Alexis Therese Petit (1791-1820), working in collaboration, found one of them. They discovered that the specific heat of elements (the temperature rise that follows upon the absorption of a fixed quantity of heat) seemed to vary inversely with the atomic weight. That is, if element x had twice the atomic weight of element y, the temperature of element x would rise by only half as many degrees as that of element y, after both had absorbed the same quantity if heat. This is the law of atomic heat.

An element with an unknown atomic weight need then only have its specific heat measured, and at once one obtains an at least rough idea as to what its atomic weight is. This method worked only for solid elements, and not for every one of them, but it was better than nothing.

Again, a German chemist, Eihardt Mitscherlich (1794-1863), had discovered, by 1819, that compounds known to have similar compositions tend to crystallize together, as though molecules of one intermingled with the similarly shaped molecules of the other.

It followed from this law of isomorphism ("same shape") that if two compounds crystallized together and if the structure of only one of them was known, the structure of the second could be assumed to be similar. This property of isomorphic crystals enabled experimenters to correct mistakes that might arise from a consideration of combining weights alone, and served as a guide to the correct atomic weights.

Weights and Symbols

The turning point came with the Swedish chemist Jons Jakob Berzelius. He, next to Dalton himself, was chiefly responsible for the establishment of the atomic theory. About 1807, Berzelius threw himself into the determination of the exact elementary constitution of various compounds. By running many hundreds of analyses, he advanced so many examples of the law of definite proportions that the world of chemistry could no longer doubt its validity and had to accept, more or less willingly, the atomic theory which had grown directly out of the law of definite proportions.

Berzelius then set about determining atomic weights with more sophistication then Dalton had been able to do. In this project, Berzelius made use of the finding of Dulong and Petit and of Mitscherlich, as well as Gay-Lussac's law of combining volumes. (He did not, however, use Avogadro's hypothesis). Berzelius's first table of atomic weights, published in 1828, compared favorably, for all but two or three elements, with the accepted values of today.

An important difference between Berzelius's table and Dalton's was that Berzelius's values were not, generally, whole numbers.

Dalton's values, based on setting the atomic weight of hydrogen equal to 1, were all given as integers. This had led the English chemist William Prout (1785-1850) to suggest, in 1815, that all the elements were, after all, but composed of hydrogen. (His suggestion at first was made anonymously). The various atoms had different weights because they were made up of different numbers of hydrogen atoms in conglomeration. This came to be called Prout's hypothesis.

Berzelius's table seemed to destroy this attractive suggestion (attractive, because it reduced the growing number of elements to one fundamental substance, after the fashion of the ancient Greeks, and thereby seemed to increase the order and symmetry of the universe). Thus, on a hydrogen-equals-1 basis, the atomic weight of oxygen is roughly 15.9, and oxygen can scarcely be viewed as being made up of fifteen hydrogen atoms plus nine-tenths of a hydrogen atom.

For the next century, better and better tables of atomic weights were published, and Berzelius's finding that the atomic weights of the various elements were not integral multiples of the atomic weight of hydrogen became clearer and clearer.

In the 1860's, for instance, the Belgian chemist Jean Servais Stas (1813-1891) determined atomic weights more accurately than Berzelius had done. Then, at the beginning of the twentieth century, the American chemist Theodore William Richards (1868-1928), taking fantastic precautions, produced atomic weight values that may represent the ultimate accuracy possible to purely chemical methods.

If Berzelius's work had left any questions, that of Stas and Richards did not. The non-integral values of the atomic weights simply had to be accepted, and Prout's hypothesis seemed deader with each stroke. Yet even as Richards was producing his remarkably precise results, the whole meaning of atomic weight had to be re-evaluated, and Prout's hypothesis rose from its ashes.

The fact that atomic weights of the different elements were not simply related also brought up the question of the proper standard against which to measure the weight. Setting the atomic weight of hydrogen equal to 1 certainly seemed natural, and both Dalton and Berzelius tried it. Still this standard gave oxygen the uneven and inconvenient atomic weight of 15.9. It was oxygen, after all, that was usually used in determining the proportions in which particular elements combined, since oxygen combined easily with so many different elements.

To give oxygen a convenient integral atomic weight with minimum interference to the hydrogen = 1 standard, its weight was shifted from 15.9 to 16.0000. On this oxygen = 16 standard, the atomic weight of hydrogen was equal, roughly, to 1.008. The oxygen = 16 standard was retained till the mid-twentieth century, when a more logical one, making very slight changes in atomic weight, was accepted.

Once the atomic theory was accepted, one could picture substances as composed of molecules containing a fixed number of atoms of various elements. It seemed very natural to try to picture such molecules by drawing the correct number of little circles, each type of atom represented by a specific type of circle.

Dalton tried this symbolism. He let a simple circle represent an oxygen atom; one with a central dot a hydrogen atom; one with a vertical line a nitrogen atom; one that was solidly black, a carbon atom, and so on. Because it becomes difficult to think up sufficiently distinct circles for each element, Dalton let some be indicated by an appropriate letter. Thus sulfur was a circle containing an "S", phosphorus one that contained a "P", and so on.

Berzelius saw that the circles were superfluous and that the initials alone would do. He suggested, therefore, that each element possess a symbol standing both for the element generally and for a single atom of that element, and that his symbol consist primarily of the initial of the Latin name of the element. (Fortunately for English-speaking people, the Latin name is almost always very like the English name). Where two or more elements possess the same initial, a second letter from the body of the name night be added. These came to be the chemical symbols of the elements, and are today internationally agreed upon and accepted.

Thus, the chemical symbols of carbon, hydrogen, oxygen, nitrogen, phosphorus, and sulfur are C, H, O, N, P, and S, respectively. The chemical symbols of calcium and chlorine (with carbon pre-empting the simple capital) are Ca and Cl, respectively. Only where the Latin names differ from the English are the symbols less than obvious. Thus, the chemical symbols for gold, silver and lead are Au ("aurum"), Ag ("argentum"), and Pb ("plumbum"), respectively.

It is easy to use these symbols to indicate the number of atoms in a molecule. If the hydrogen molecule is made up of two atoms of hydrogen, it is H2. If the water molecule contains two atoms of hydrogen and one of oxygen, it is H2O. (The symbol without a number represents a single atom.) Again, carbon dioxide is CO2 and sulfuric acid is H2SO4, while hydrogen chloride is HCl. The chemical formulas of these simple compounds are self-explanatory.

Chemical formulas can be combined to form a chemical equation and describe a reaction. If one wishes to express the fact that carbon combines with oxygen to form carbon dioxide, one can write:

C + O2 --> CO2

Such equations must account for all the atoms if Lavoisier's law of conservation of mass is to be obeyed. In the equation just cited, for instance, you begin with an atom of C (carbon) and two atoms of O (the oxygen molecule), and you end with an atom of C and two atoms of O (the carbon dioxide molecule).

Suppose, though, you wished to say that hydrogen combined with chlorine to form hydrogen chloride. If this were written to simply

H2 + Cl2 --> HCl,

it could be pointed out that there were two atoms of hydrogen and two atoms of chlorine, to begin with, but only one of each at the conclusion. To write a balanced chemical equation, one must say:

H2 + Cl2 --> 2HCl.

In the same way, to describe the combination of hydrogen and oxygen to form water, we can write a balanced equation:

2H2 + O2 --> 2H2O

Electrolysis

Meanwhile, the electric current, which had been used to such good effect by Nicholson and Carlisle, produced even more startling effects in the isolation of certain new elements.

Since Boyle's definition of "element" a century and a half before, substances qualifying as elements by that definition were discovered in astonishing numbers. More frustratingly, some substances were known which were not elements, yet contained undiscovered elements that chemists could not manage to study in isolation.

Thus, elements are frequently found in combination with oxygen (as oxides). To free the element it was necessary to remove the oxygen. If a second element with a stronger affinity for oxygen were to be introduced, perhaps the oxygen would leave the first element and become attached to the second. The method was found to work. Often carbon did the trick. Thus iron ore, which is essentially iron oxide, could be heated with coke (a relatively pure form of carbon). The carbon would combine with the oxygen to form carbon monoxide and carbon dioxide, and metallic iron would be left behind.

But now consider lime instead. From its properties lime, too, seems to be an oxide. However, no known element forms lime on combination with oxygen, and one must conclude that lime is a compound of an unknown element with oxygen. To isolate that unknown element, one might try to heat lime with coke; but if so, nothing happens. The unknown element hold oxygen so strongly that carbon atoms are powerless to snatch the oxygen atoms away. Nor could any other chemical strip lime of its oxygen.

It occurred to an English chemist, Humphry Davy (1778-1829), that what could not be pulled apart by chemicals might be forced apart by the strange power of the electric current, which could pry apart the water molecule with ease when chemicals were helpless.

Davy proceeded to construct an electric battery with over 250 metallic plates, the strongest ever built up to that time. He ran intense currents from this battery through solutions of the compounds suspected of containing unknown elements, but did so without result. He obtained only hydrogen and oxygen from the water.

Apparently, he had to eliminate water. However, when he used the solid substances themselves, he could not make a current pass through them. It occurred to him, finally, to melt the compounds and pass the current through the melt. He would then, so to speak, be using a waterless, conducting liquid.

This scheme worked. On October 6, 1807, Davy passed a current through molten potash (potassium carbonate) and liberated little globules of a metal he at once labeled potassium. (It was so active it pulled oxygen away from water, liberating hydrogen with enough energy to cause it to burst into flame). A week later, Davy isolated sodium from soda (sodium carbonate), an element only slightly less active than potassium.

In 1808, by using a somewhat modified method suggested by Berzelius, Davy isolated several metals from their oxides; magnesium from magnesia, strontium from strontia, barium from baryta, and calcium from lime. ("Calcium" is from the Latin word for lime.)

Among other things, Davy also showed that a certain greenish-gas, which Scheele had discovered a generation earlier and thought to be an oxide, was actually an element. Davy suggested the name chlorine, from the Greek word for "green". Davy also showed that hydrochloric acid, although a strong acid, contained no oxygen atom in its molecule, this disproving Lavoisier's suggestion that oxygen was a necessary component of acids.

Davy's work on electrolysis was extended by his assistant and protege, Michael Faraday (1791-1867), who grew to be an even greater scientist then his teacher. Faraday, in working with electrochemistry, introduced a number of terms that are still used today. It was he, for instance, who first termed the splitting of molecules by an electric current, electrolysis. At the suggestion of the English classical scholar William Whewell (1794-1866) Faraday named a compound or solution which could carry an electric current, an electrolyte. The metal rods or strips inserted into a melt or solution, he called electrodes; the electrode carrying a positive charge being an anode, the one carrying a negative charge the cathode.

The electric current was carried through the melt or solution by entities Faraday called ions (from a Greek word meaning "wanderer"). Those ions that traveled to the anode he called anions; those that traveled to the cathode were cations.

In 1832, he was able to announce the existence of certain quantitative relationships in electrochemistry. His first law of electrolysis stated: The mass of substance liberated at an electrode during electrolysis is proportional to the quantity of electricity driven through the solution. His second law of electrolysis stated: The weight of metal liberated by a given quantity of electricity is proportional to the equivalent weight of the metal.

Thus, if 2.7 times as much silver as potassium will combine with a given quantity of oxygen, then 2.7 times as much silver as potassium will be liberated from its compounds by a given quantity of electricity.

Faraday's law of electrolysis seemed to indicate, in the view of some chemists, that electricity could be subdivided into fixed, minimum units, as matter itself could. In other words, there were "atoms of electricity".

Suppose that when electricity passed through a solution, atoms of matter were dragged to either the cathode or the anode by "atoms of electricity". Suppose that, often, one "atom of electricity" sufficed to handle one atom of matter, but that sometimes two or even three "atoms of electricity" were required. In that case Faraday's laws of electrolysis could easily be explained.

It was not until the very end of the nineteenth century that this view was established and the "atoms of electricity" were located. Faraday, himself, however, was never enthusiastic about "atoms of electricity" or, indeed, about atomism in general.