Illustrating Atoms and Molecules
This article appears in the 2011 Journal of Natural Science Illustration
Abstract: Since atoms are smaller than the wavelength of visible light, it is theoretically impossible to “see” an atom, even with the most powerful microscope. Nevertheless, we recognize that atoms consist of “shells” of electrons buzzing around a central nucleus. Therefore, it’s common to depict an atom as a simple sphere, its diameter proportional to the size of its outermost electron shell. Furthermore, scientists have developed experimental methods, such as x-ray crystallography and NMR spectroscopy, to determine the geometric arrangement of atoms within a molecule. These data can be used to construct three-dimensional models of molecules, but the illustrator must be aware that such a model is an abstract representation and is not meant to show what the molecule really “looks like”.
Because an atom is smaller than the wavelength of visible light, it cannot reflect light and, therefore, has no color. The colorful atoms you see in chemistry textbooks are based on conventions that have been adopted by chemists over several centuries. The alchemists of the Middle Ages and Renaissance used iconic symbols to depict the different elements (Figure 1). They also associated certain colors with each element based on its physical properties, although these colors never appeared in print because of the rarity of color printing prior to the late 19th Century.
By the mid-1800s, it was recognized that atoms bonded to one another to form molecules. In 1865, the chemist August Hoffman gave a Friday Evening Discourse at London’s Royal Institution on the “Combining Power of Atoms.” In order to demonstrate chemical bonding of atoms, he drilled holes in croquet balls and connected them with metal pipes (Figure 2). His choices were limited by the available colors of croquet balls and he relied on many of the same color conventions that had been adopted centuries earlier:
- Carbon is colored black because it’s the color of charcoal.
- Oxygen is red because it’s necessary for combustion.
- Nitrogen is blue because it’s the most abundant element in the Earth’s atmosphere and the sky appears blue.
- Hydrogen is white because it forms a colorless gas.
- Chlorine is green because it forms a greenish gas.
- Sulfur is yellow because that’s its color in mineral form.
- Phosphorus is orange because it glows orange in a flame.
- Iron is reddish brown because it rusts.
With minor variations, these color conventions are still in use today. In the 1950s, Robert Corey and Linus Pauling at CalTech developed a set of wooden atomic models for constructing molecular models. Soon thereafter, plastic molecular model kits became popular in chemistry classes. The Corey-Pauling models were further refined by Walter Koltun at NIH in the early 1960s. The color scheme used in these models is now known as “CPK” after Corey, Pauling, and Koltun. See Wikipedia for a complete list of CPK colors of the elements: http://en.wikipedia.org/wiki/CPK_coloring
In 1945, Linus Pauling described the van der Waal’s radius, a measure of the size of an atom. Strictly speaking, it represents “one-half the distance between two equivalent nonbonded atoms in their most stable arrangement.” Put another way, it is the closest that two unbonded atoms can get before they begin to repel one another. However, most people think of it simply as the size of the atom’s outermost electron shell.
Many molecular models use the van der Waal’s radius to distinguish the relative size of different atoms and to give a sense of the overall size and shape of the molecule. Figure 3 shows the van der Waal’s radii of the elements arranged in a simplified periodic table (it also shows the CPK colors of common elements). Note that hydrogen and helium in the top row are tiny because they have just a single shell with only one or two electrons. As you go down the table, from top-to-bottom, the elements in each row are larger because each row adds an additional electron shell. As you go from left-to-right across the table, the elements in each column become somewhat smaller. This is surprising because each column represents the addition of one more proton to the nucleus and one more electron to the outermost shell. You might assume that this would make the elements on the right side larger than those on the left. However, the atoms on the right are smaller because there is a greater attraction between the positively charged nucleus and negative electrons, causing the electrons to orbit closer to the nucleus.
In 1858, the English chemist Archibald Couper published the first drawing of a molecule, using simple lines to connect atoms to one another. In the same year, Freidrich August Kekulé von Stradonitz published what would later be known as the theory of valence, that atoms of each element form a specific number of bonds (e.g., carbon always forms four bonds, nitrogen forms three, oxygen forms two, etc.). In 1874, Jacobus van ’t Hoff, a former student in Kekulé’s lab, demonstrated that these bonds are arranged in specific geometric configurations. For example, the four bonds of a carbon atom are arranged in a tetrahedron, spaced apart at equal 109.5° angles. Since then, it has been possible to construct 3D models of simple molecules just by knowing which atoms are bonded to one another. (In larger molecules, such as proteins and DNA, there are other forces at work and the shape cannot be predicted from bond angles alone.)
Molecular visualization software now makes it easy to create accurate 3D models of any molecule. Figure 4 shows three common methods for representing simple molecules. A stick model shows only the bonds and does not show the atoms themselves. A ball-and-stick model adds small spheres to represent the center of each atom. A space-filling model uses much larger spheres proportionate in size to each atom’s van der Waal’s radius. Space-filling models were first developed by Robert Corey and Linus Pauling and later refined by Walter Koltun. Therefore, the acronym “CPK” may be used to describe either the space-filling style or the color scheme used for the individual atoms. Figure 5 shows an additional example of space-filling models.
I am often asked which type of representation is the best for illustrating molecules. As is often the case in scientific illustration, the answer depends on what you are trying to show. Many illustrators prefer space-filling models because the large spheres lend themselves to dramatic lighting effects (see Figure 6). However, one must realize that these effects are purely artistic since an atom is incapable of producing a highlight, a core shadow, or reflections. More importantly, a space-filling model completely obscures the bonds between atoms, making it nearly impossible to tell which atom is connected to which. Therefore, space-filling models are ideal for editorial illustration but may not be suitable for a chemistry textbook where it’s necessary to see the bonding between atoms.
Because of its central importance to biology, scientific illustrators are often called upon to illustrate the DNA molecule. Unfortunately, many illustrations of DNA, even those in science textbooks and scientific websites, are inaccurate. In fact, entire websites have been created to catalog inaccurate DNA images, e.g., the Left-Handed DNA Hall of Fame: http://www-lmmb.ncifcrf.gov/~toms/LeftHanded.DNA.html
All nucleic acids, including DNA and RNA, are formed from subunits called nucleotides. Each nucleotide consists of a base, plus a sugar molecule, plus a phosphate. In DNA, the bases are adenine, guanine, cytosine, and thymine (see Figure 7). Uracil is substituted for thymine in RNA. After addition of a sugar molecule, each base is called a nucleoside (Figure 8). In RNA, the sugar is called ribose. In DNA, it is deoxyribose. Adding a phosphate to the 5’ carbon of the sugar molecule creates a nucleotide.
Nucleotides, in turn, are linked into long strands by creating bonds between the phosphate and 3’ carbon of the sugar molecule (Figure 9). In DNA, two complementary strands are joined by hydrogen bonds between the bases. The two strands resemble a ladder with the sugar-phosphates forming the sides of the ladder (or “backbone”) and the bases facing in to form the “rungs” of the ladder.
The DNA “ladder” is twisted to form the distinctive double helix. Figure 10 is a schematic but highly accurate representation of the double helix, showing the precise measurements (in Angstroms) of each part of the helix. Note that there are ten pairs of bases for each complete turn (wavelength or “pitch”) of the helix. Because of the way the ladder twists, it forms distinct major and minor grooves. The major groove is exactly twice the width of the minor groove.
Failing to distinguish the two grooves is one of the most common errors in illustrating DNA. However, by far the most common error is reversing the handedness of the helix. Every spiral can be described as either left-handed or right-handed (Figure 11). For unknown reasons, right-handed spirals are much more common in nature, as well as in man-made spirals such as threads on a screw. Although a left-handed form of DNA does exist (zDNA), the naturally occurring form found in the nucleus of living organisms (called bDNA) is always a right-handed helix.
Left- and right-handed spirals are mirror images of one another. Because computer graphics software makes it so easy to flip an image, it is very easy to inadvertently reverse the handedness of a DNA helix. I suspect one reason for the prevalence of left-handed DNA images is that art directors unwittingly flip the art for purely aesthetic reasons without realizing that it affects the scientific content of the illustration.
DNA can be represented using any of the styles described above (stick, ball-and-stick, space-filling). Figure 12 shows a space-filling model with CPK colors rendered in Cinema 4D. In addition, DNA is often represented as a ladder (as in Fig. 10) or using icons to represent the nucleotide bases. Figure 12 shows a 3D model of DNA with such icons. This figure also uses a color scheme to differentiate the bases. Although not as universal as the CPK color scheme, this system is used by some molecular visualization software and by the online Nucleic Acid Database. The system uses the first letter of each base name to determine its color:
- Adenine (A) = azure (blue)
- Guanine (G) = green
- Cytosine (C) = crimson (red)
- Thymine (T) = “Tweety Bird” (yellow)
- Uracil (U) = umber (brown).
About the Author
Jim Perkins is Professor of Medical Illustration in the College of Health Sciences and Technology at Rochester Institute of Technology. Prof. Perkins is a Board Certified Medical Illustrator (CMI), Fellow of the Association of Medical Illustrators (FAMI), and currently serves as President of the Vesalius Trust for Visual Communication in the Health Sciences. An expert in the visual communication of complex biomedical subject matter, particularly in the areas of cell biology, molecular biology, physiology, and pathology, he has illustrated over 40 medical textbooks and serves as a consultant to major medical publishers.
Prof. Perkins received his Bachelor’s degree in Biology and Geology from Cornell University (1985), and studied paleontology and anatomy at the University of Texas, Austin and University of Rochester, completing his PhD coursework (1989). He received an MA in Medical Illustration from RIT (1992), and following work in medical publishing and the medical legal exhibit field, he joined the RIT faculty in 1998.