Visual walkthrough — Group 14 (Carbon family) — allotropes of C (diamond, graphite, fullerenes, graphene, CNTs)

We are answering one question: when carbon uses $s{p}^3$ hybrid orbitals to make four identical bonds, how far apart (in angle) does each pair of bonds sit? See Hybridisation sp sp2 sp3 for why the orbitals are identical; here we take that as given and compute the geometry.

---

Step 1 — What "bond angle" even means

WHAT: We draw one central carbon and two of its bonds as arrows pointing away from it.

WHY: Before computing anything, we must agree that "the angle" lives at the carbon, between two bond directions. Everything below is just finding those two directions precisely.

PICTURE: The two orange arrows leave the navy carbon; the shaded wedge between them is the quantity $\theta$ we want.

---

Step 2 — Why a cube? The symmetry trick

WHAT: Draw a cube. Colour 4 of its 8 corners so that no two coloured corners share an edge — they are the diagonal corners. Put a carbon at the cube's centre.

WHY this tool (a cube)? Because a cube's corners have coordinates that are just $\pm 1$ in each direction — the simplest numbers imaginable. Any harder shape would give ugly coordinates. The cube hands us the four bond directions for free.

PICTURE: The 4 magenta corners are the tetrahedron; the 4 faded corners are ignored. Notice each magenta corner is "surrounded" by faded ones — that spacing is the tetrahedral spread.

---

Step 3 — Naming the corners with numbers (coordinates)

WHAT: Centre the cube on the origin so its 8 corners are all combinations of $\pm 1$. The four alternate (tetrahedral) corners work out to:

$\vec{a}=(+1,+1,+1),\vec{b}=(+1,-1,-1)$ $\vec{c}=(-1,+1,-1),\vec{d}=(-1,-1,+1)$

• $\vec{a}$ :: the corner where all three addresses are positive — "up, right, toward you".
• $\vec{b}$ :: flip two signs — a corner that shares no edge with $\vec{a}$.
• The pattern: each of $\vec{a},\vec{b},\vec{c},\vec{d}$ has an even number of minus signs (0 or 2), which is exactly the "no shared edge" rule.

WHY: To compute an angle we need the bond directions as numbers. These four arrows, from centre to corner, ARE the four bond directions.

PICTURE: Each magenta corner is labelled with its $(\pm 1,\pm 1,\pm 1)$ address; count the minus signs — always 0 or 2.

---

Step 4 — The right tool for "angle between two arrows": the dot product

We have two arrows $\vec{a}$ and $\vec{b}$. We want the angle between them. Which mathematical tool answers "angle between two arrows"? The dot product.

WHY this tool and not another? The dot product is the only simple operation that takes two direction-arrows and returns a single number that depends only on the angle between them (once we account for their lengths). Its two faces — one you can compute from coordinates, one that contains the angle — let us solve for $\theta$.

Reading each symbol:

• $\vec{a}\cdot \vec{b}$ :: the single number we build from the two arrows.
• $a_1{b}_1+a_2{b}_2+a_3{b}_3$ :: multiply matching addresses, add them up — pure arithmetic.
• $|\vec{a}|$ :: the length of arrow $\vec{a}$ (how long from centre to corner).
• $\cos \theta$ :: cosine, the "how aligned are they?" dial — $\cos 0^{\circ }=1$ (same direction), $\cos 90^{\circ }=0$ (perpendicular), $\cos 180^{\circ }=-1$ (opposite).

PICTURE: Two arrows with the angle $\theta$; when they point the same way $\cos \theta =1$, when opposite $\cos \theta =-1$ — the dial swings from $+1$ to $-1$.

---

Step 5 — The lengths $|\vec{a}|$ and $|\vec{b}|$

WHAT: For $\vec{a}=(1,1,1)$: $|\vec{a}|=\sqrt{1^2+1^2+1^2}=\sqrt{3}$ and identically $|\vec{b}|=\sqrt{1^2+(-1{)}^2+(-1{)}^2}=\sqrt{3}$.

• $1^2$ :: squaring kills the minus signs, so it doesn't matter which corner — every one is distance $\sqrt{3}$ from centre.

WHY: The geometry side of the dot product carries the factor $|\vec{a}||\vec{b}|$. We must know it to isolate $\cos \theta$. Because all four arrows have the same length $\sqrt{3}$, the setup is beautifully symmetric — confirming these really are four identical bonds.

PICTURE: The arrow from origin to $(1,1,1)$, with the little right-angle boxes showing the two Pythagoras steps that build up $\sqrt{3}$.

---

Step 6 — Put it together and solve for $\theta$

WHAT — the arithmetic side: $\vec{a}\cdot \vec{b}=(1)(1)+(1)(-1)+(1)(-1)=1-1-1=-1$

WHAT — set it equal to the geometry side: $\vec{a}\cdot \vec{b}=|\vec{a}||\vec{b}|\cos \theta ⟹-1=\sqrt{3}\cdot \sqrt{3}\cdot \cos \theta =3\cos \theta$

WHAT — isolate the dial: $\cos \theta =\frac{-1}{3}$

• $-1$ :: the arithmetic dot product from the coordinates.
• $3$ :: the product $\sqrt{3}\cdot \sqrt{3}$ of the two lengths.
• $\cos \theta =-\frac{1}{3}$ :: the number is negative, which already tells us $\theta >90^{\circ }$ — the bonds lean away from each other, exactly what "spread out as far as possible" should look like.

Now undo the cosine. To go from "$\cos \theta$ equals this" back to "$\theta$ equals what?" we need the tool that reverses cosine: arccosine (${\cos }^{-1}$), the question "which angle has this cosine?"

$\theta ={\cos }^{-1}\left(-\frac{1}{3}\right)=109.47^{\circ }$

WHY arccos and not anything else? Cosine turns an angle into a ratio; here we have the ratio and want the angle back — that is precisely the job arccos does. It undoes cosine the way subtraction undoes addition.

PICTURE: The number line from $+1$ to $-1$ with $-\frac{1}{3}$ marked, and the arc swinging up to $109.47^{\circ }$ — past the $90^{\circ }$ mark, confirming the "leaning away" sign.

---

Step 7 — Sanity checks and the degenerate cases

We must never leave a scenario unshown. Test the same method on limits we already know:

Since $-\frac{1}{3}$ sits between $0$ (which gives $90^{\circ }$) and $-1$ (which gives $180^{\circ }$), we knew before computing that $109.47^{\circ }$ must land between $90^{\circ }$ and $180^{\circ }$ — and it does.

---

The one-picture summary

Everything above compressed: a cube, four magenta body-diagonal bonds from its centre, two of them highlighted with the wedge $\theta ={\cos }^{-1}(-1/3)=109.47^{\circ }$, and the three-line derivation printed beside it.

Recall Feynman retelling — explain it to a friend

We wanted to know how far apart diamond's four bonds sit. Trick: draw a cube and send the four bonds from its middle out to four corners that don't touch each other — that's the most spread-out, most symmetric arrangement, and it's what "tetrahedral" means. Those corners have dead-simple addresses like $(1,1,1)$ and $(1,-1,-1)$. To find the angle between two arrows we use the dot product, which has two personalities: one you compute from the numbers ($1-1-1=-1$) and one that hides the angle inside ($3\cos \theta$). Set them equal, get $\cos \theta =-1/3$. The minus sign already whispers "more than $90^{\circ }$ — the bonds lean apart." Ask "which angle has that cosine?" (arccos) and out pops $109.47^{\circ }$. We checked the method on easy cases — opposite arrows give $180^{\circ }$, perpendicular give $90^{\circ }$ — so we trust it. That single number is why diamond's atoms lock into a rigid 3-D web, and hence why it's the hardest thing you'll ever hold.

---

Connections

• Parent: Group 14 — Allotropes of Carbon
• Hybridisation sp sp2 sp3
• Giant covalent vs molecular solids
• Silicon and its differences from carbon