Trigonometric functions

Sine and cosine start life as ratios in a right triangle, useful only for acute angles below 90∘90^\circ. Trigonometric functions break that ceiling. By reading the coordinates of a point moving around a circle of radius 11, sine and cosine become functions of any real number — negative, zero, or larger than a full turn — and the other four ratios follow from them. This single idea, the unit-circle definition, is the spine of the whole chapter: it delivers the fundamental identity sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 for free, fixes the exact domain and range of all six functions, and explains why some of them repeat every 2π2\pi while others repeat every π\pi. Everything that follows — signs in each quadrant, evaluating sin⁡(7π/6)\sin(7\pi/6), solving equations — is this one picture used again and again.

The core insight

Defining cos⁡x\cos x and sin⁡x\sin x as the coordinates of a point reached by travelling an arc length xx around the unit circle extends trigonometry from acute-angle ratios to functions of every real number, with sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 following directly from the circle's equation.

The core idea

From a triangle ratio to a circle picture. In a right triangle, sin⁡θ\sin\theta and cos⁡θ\cos\theta only make sense for 0∘<θ<90∘0^\circ<\theta<90^\circ. To go beyond that, take a circle of radius 11 centred at the origin OO — the unit circle — and a starting point P0=(1,0)P_0=(1,0) on it. For a real number xx, walk along the circle through an arc length xx, anticlockwise if x>0x>0 and clockwise if x<0x<0, to land on a point P(a,b)P(a,b). Define

cos⁡x=a,sin⁡x=b.\cos x = a, \qquad \sin x = b.

where: xx is the arc length travelled, measured in radians (so xx doubles as the angle ∠P0OP\angle P_0OP), and (a,b)(a,b) are the coordinates of the point reached. Because the radius is 11, arc length and radian angle are the same number, so this matches the old right-triangle sine and cosine for acute xx while also making sense for x=0x=0, negative xx, or xx bigger than 2π2\pi (multiple loops around the circle). The remaining four functions are ratios built from sine and cosine, defined wherever the denominator does not vanish:

tan⁡x=sin⁡xcos⁡x (cos⁡x≠0),cot⁡x=cos⁡xsin⁡x (sin⁡x≠0),\tan x = \frac{\sin x}{\cos x}\ (\cos x\neq0), \qquad \cot x = \frac{\cos x}{\sin x}\ (\sin x\neq0),
sec⁡x=1cos⁡x (cos⁡x≠0),cosec⁡x=1sin⁡x (sin⁡x≠0).\sec x = \frac{1}{\cos x}\ (\cos x\neq0), \qquad \operatorname{cosec} x = \frac{1}{\sin x}\ (\sin x\neq0).

where: each function is undefined exactly where its denominator is zero — a fact worth memorising because it fixes the domain of four of the six functions.

Figure
A unit circle with a point P at arc length x showing cosine as the horizontal coordinate and sine as the vertical coordinate
The unit circle definition of sine and cosine P0 = (1,0) P(a,b) arc length x b = sin x a = cos x P moves anticlockwise for positive x; its coordinates are cos x and sin x

Domain, range, and period. Because a2+b2=1a^2+b^2=1 on the unit circle, both sin⁡x\sin x and cos⁡x\cos x take every value in [−1,1][-1,1] and are defined for all real xx — there is no gap or forbidden point to worry about for these two. tan⁡x\tan x and sec⁡x\sec x break down wherever cos⁡x=0\cos x=0, that is at x=(2n+1)π2x=(2n+1)\tfrac{\pi}{2}; cot⁡x\cot x and cosec⁡x\operatorname{cosec} x break down wherever sin⁡x=0\sin x=0, that is at x=nπx=n\pi (n∈Zn\in\mathbb{Z}). Their ranges are R\mathbb{R} for tangent and cotangent, and R−(−1,1)\mathbb{R}-(-1,1) for secant and cosecant, since dividing 11 by a number in [−1,1][-1,1] (excluding 00) always gives something of size at least 11: as cos⁡x\cos x shrinks toward 00, sec⁡x=1/cos⁡x\sec x=1/\cos x grows without bound, so no value strictly between −1-1 and 11 can ever occur. Sine and cosine (and therefore secant and cosecant) repeat every 2π2\pi, one full lap of the circle; tangent and cotangent, because a half-turn sends both coordinates to their negatives and their ratio is unchanged, already repeat after just π\pi, half a lap — the shortest interval after which every one of the six functions can be guaranteed to return to its starting value is therefore 2π2\pi, but tangent and cotangent settle back sooner.

Worked example

Problem. If cos⁡x=−35\cos x=-\dfrac{3}{5} and xx lies in the third quadrant, find sin⁡x\sin x, tan⁡x\tan x, and, as a stretch, sec⁡x\sec x, cosec⁡x\operatorname{cosec} x, and cot⁡x\cot x.

Step 1. Use the fundamental identity to get sin⁡2x\sin^2x: sin⁡2x=1−cos⁡2x=1−925=1625\sin^2x=1-\cos^2x=1-\dfrac{9}{25}=\dfrac{16}{25}, so sin⁡x=±45\sin x=\pm\dfrac{4}{5}.

Step 2. Fix the sign from the quadrant. In the third quadrant both sine and cosine are negative (only tangent and cotangent are positive there), matching the given negative cos⁡x=−35\cos x=-\dfrac{3}{5}. So take the negative root: sin⁡x=−45\sin x=-\dfrac{4}{5}.

Step 3. Compute tan⁡x=sin⁡xcos⁡x=−4/5−3/5=43\tan x=\dfrac{\sin x}{\cos x}=\dfrac{-4/5}{-3/5}=\dfrac{4}{3}.

Competitive stretch

Step 4. The remaining three functions are reciprocals of what we already have: sec⁡x=1cos⁡x=−53\sec x=\dfrac{1}{\cos x}=-\dfrac{5}{3}, cosec⁡x=1sin⁡x=−54\operatorname{cosec} x=\dfrac{1}{\sin x}=-\dfrac{5}{4}, and cot⁡x=1tan⁡x=34\cot x=\dfrac{1}{\tan x}=\dfrac{3}{4}. As a quick check, 1+tan⁡2x=1+169=259=sec⁡2x1+\tan^2x=1+\dfrac{16}{9}=\dfrac{25}{9}=\sec^2x, which matches (−53)2=259\left(-\dfrac{5}{3}\right)^2=\dfrac{25}{9}.

Final answer: sin⁡x=−45\sin x=-\dfrac{4}{5}, tan⁡x=43\tan x=\dfrac{4}{3}, sec⁡x=−53\sec x=-\dfrac{5}{3}, cosec⁡x=−54\operatorname{cosec} x=-\dfrac{5}{4}, cot⁡x=34\cot x=\dfrac{3}{4}.

Derivation AThe fundamental identity from the unit circle

The goal is to prove, using nothing beyond the equation of a circle and the unit-circle definition of sine and cosine, that sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 holds for every real number xx, and then to turn that one identity into the two companion identities 1+tan⁡2x=sec⁡2x1+\tan^2x=\sec^2x and 1+cot⁡2x=cosec⁡2x1+\cot^2x=\operatorname{cosec}^2x by simple division, so a student can see all three come from a single geometric fact rather than three separate rules to memorise.

Step 1. Write down the equation of the unit circle. Every point (a,b)(a,b) on a circle of radius 11 centred at the origin satisfies a2+b2=1a^2+b^2=1 — this is just the Pythagorean theorem applied to the right triangle formed by the point, the origin, and the foot of the perpendicular to the horizontal axis, with hypotenuse equal to the radius 11.

Step 2. Substitute the unit-circle definition. By definition, the point PP reached after arc length xx has coordinates (a,b)=(cos⁡x,sin⁡x)(a,b)=(\cos x,\sin x). Substituting these into a2+b2=1a^2+b^2=1 gives immediately

sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1

for every real xx, since the construction of PP places it on the unit circle no matter how large or small xx is, or which direction it is measured.

Step 3. Divide by cos⁡2x\cos^2x to get the tangent–secant identity. Wherever cos⁡x≠0\cos x\neq0, dividing every term of sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 by cos⁡2x\cos^2x gives sin⁡2xcos⁡2x+1=1cos⁡2x\dfrac{\sin^2x}{\cos^2x}+1=\dfrac{1}{\cos^2x}, which is exactly tan⁡2x+1=sec⁡2x\tan^2x+1=\sec^2x, using the ratio definitions of tan⁡x\tan x and sec⁡x\sec x from above.

Step 4. Divide by sin⁡2x\sin^2x to get the cotangent–cosecant identity. Wherever sin⁡x≠0\sin x\neq0, dividing every term of sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 by sin⁡2x\sin^2x gives 1+cos⁡2xsin⁡2x=1sin⁡2x1+\dfrac{\cos^2x}{\sin^2x}=\dfrac{1}{\sin^2x}, that is 1+cot⁡2x=cosec⁡2x1+\cot^2x=\operatorname{cosec}^2x. All three identities are thus one statement — the Pythagorean theorem on the unit circle — viewed three ways.

Common traps

  • Guessing the sign instead of using the quadrant. Taking a square root from sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 always gives two possible signs; the quadrant stated in the problem, not a default choice of positive, decides which one is correct.
  • Ignoring domain restrictions. tan⁡x\tan x and sec⁡x\sec x are undefined at odd multiples of π2\dfrac{\pi}{2}; cot⁡x\cot x and cosec⁡x\operatorname{cosec} x are undefined at multiples of π\pi. Plugging these values in anyway is a common slip.
  • Mixing up the two periods. Sine and cosine repeat every 2π2\pi, but tangent and cotangent repeat every π\pi — using 2π2\pi where π\pi is correct throws off simplification and equation-solving questions.
  • Treating the range of secant and cosecant as excluding just zero. Their range excludes the entire open interval (−1,1)(-1,1), not only the single point 00.

Ask the tutor: want to see the unit-circle picture animated, or practice evaluating a few more angles like 7π6\dfrac{7\pi}{6} or 5π3\dfrac{5\pi}{3} step by step?

Why does the unit circle need radius exactly 1?

Radius 11 is what makes arc length and radian angle the same number, so a single quantity xx can serve as both "how far along the circle" and "the angle turned through." With any other radius rr, the arc length for angle xx radians would be rxrx, and the definitions cos⁡x=a\cos x=a, sin⁡x=b\sin x=b would need an extra division by rr to stay consistent. Choosing radius 11 removes that clutter and keeps the definitions clean.

Why do tan x and cot x repeat every pi instead of every 2 pi like sine and cosine?

Moving halfway around the circle, from xx to x+πx+\pi, sends the point P(a,b)P(a,b) to its exact opposite (−a,−b)(-a,-b) — both coordinates flip sign. So sin⁡(x+π)=−sin⁡x\sin(x+\pi)=-\sin x and cos⁡(x+π)=−cos⁡x\cos(x+\pi)=-\cos x, meaning sine and cosine individually need a full 2π2\pi to return to their original values. But tan⁡(x+π)=−sin⁡x−cos⁡x=sin⁡xcos⁡x=tan⁡x\tan(x+\pi)=\dfrac{-\sin x}{-\cos x}=\dfrac{\sin x}{\cos x}=\tan x: the two sign flips cancel in the ratio, so tangent (and by the same argument cotangent) already returns to its original value after just π\pi.

How do I quickly find sin, cos and tan of angles like 7 pi over 6 without a calculator?

Split the angle into a multiple of π\pi (or π/2\pi/2) plus a small standard angle, then use the sign rule for that quadrant together with the value of the standard angle. For 7π6=π+π6\dfrac{7\pi}{6}=\pi+\dfrac{\pi}{6}, the angle π+θ\pi+\theta lands in the third quadrant, where sine and cosine are both negative, and the standard results sin⁡(π+θ)=−sin⁡θ\sin(\pi+\theta)=-\sin\theta, cos⁡(π+θ)=−cos⁡θ\cos(\pi+\theta)=-\cos\theta let you read off sin⁡7π6=−12\sin\dfrac{7\pi}{6}=-\dfrac12 and cos⁡7π6=−32\cos\dfrac{7\pi}{6}=-\dfrac{\sqrt3}{2} straight from the known values at θ=π/6\theta=\pi/6.

Before this

Where this leads