Exponential Functions & Laws of Exponents

⚡ Key Takeaways

Exponential functions have the form $f(x) = a \cdot b^x$ , where the variable sits in the exponent. The laws of exponents are algebraic rules ( $b^m b^n = b^{m+n}$ , etc.) that govern how powers combine. These functions model anything with a constant percentage rate of change — growth (populations, money, viruses) and decay (radioactivity, cooling, drug clearance). Codified gradually from Archimedes through Napier, Euler, and the 17th–18th centuries.

💡 The Core Insight

Every law of exponents is a consequence of a single principle: a power $b^n$ counts $n$ copies of $b$ multiplied together, so combining powers of the same base just means counting copies. Multiplying $b^m \cdot b^n$ stacks $m$ copies next to $n$ copies, giving $m+n$ copies total. Once we insist this addition rule keeps working even when $n$ is zero, negative, or fractional, the definitions of $b^0$ , $b^{-n}$ , and $b^{1/n}$ are forced — they are not arbitrary conventions but the only choices consistent with the master rule.

DefinitionExponential Function

A function $f : \mathbb{R} \to \mathbb{R}$ is exponential if it has the form

1

\quad f(x) = a \cdot b^x, \qquad a \neq 0,\ b > 0,\ b \neq 1.

Derivation hint: the defining feature is that the input variable $x$ appears in the exponent, not the base; contrast with a power function $x^n$ . Here:

$b$ is the base (the constant growth/decay factor per unit increase in $x$ ),
$a$ is the initial value, since $f(0) = a \cdot b^0 = a$ ,
$x$ is the variable exponent.

We require $b > 0$ so that $b^x$ is defined for all real $x$ (e.g. $(-1)^{1/2}$ is not real), and $b \neq 1$ so the function is not the constant $f(x) = a$ .

DefinitionLaws of Exponents

For real bases $b, c > 0$ and any real exponents $m, n$ :

2

\quad b^{m} \cdot b^{n} = b^{\,m+n}

Derivation hint: count $m$ factors of $b$ followed by $n$ factors of $b$ , giving $m+n$ factors.

3

\quad \frac{b^{m}}{b^{n}} = b^{\,m-n}

Derivation hint: dividing cancels $n$ of the factors from the numerator; equivalently apply [2] with exponent $m-n$ then add $n$ .

4

\quad (b^{m})^{n} = b^{\,m n}

Derivation hint: raising to the $n$ th power repeats the block of $m$ factors $n$ times, giving $mn$ factors.

5

\quad (b \, c)^{n} = b^{n} \, c^{n}

Derivation hint: regroup $n$ copies of $(bc)$ by commutativity into $n$ copies of $b$ times $n$ copies of $c$ .

6

\quad b^{0} = 1, \qquad b^{-n} = \frac{1}{b^{n}}, \qquad b^{1/n} = \sqrt[n]{b}

Derivation hint: each is forced by demanding [2] still hold — e.g. $b^n b^0 = b^{n+0}=b^n \Rightarrow b^0=1$ .

TheoremCharacterization by Constant Ratio

A function $f$ defined on the integers with $f(x+1)/f(x) = b$ constant for all $x$ is exponential: $f(x) = f(0)\, b^x$ . Over the reals, the continuous solutions of $f(x+y) = f(x)f(y)$ with $f$ not identically zero are exactly $f(x) = b^x$ for some $b > 0$ .

ProofDeriving the zero, negative, and fractional exponents from Law 2

We take Law [2], $b^m b^n = b^{m+n}$ , as the master rule and force consistency.

Step 1 — Zero exponent. Set $n = 0$ in [2]:

7

\quad b^{m}\cdot b^{0} = b^{m+0} = b^{m}.

Derivation hint: substitute $n=0$ directly into [2]. Dividing both sides by $b^m$ (legal since $b \neq 0$ ) gives $b^0 = 1$ .

Step 2 — Negative exponent. Set $n = -m$ in [2]:

8

\quad b^{m}\cdot b^{-m} = b^{m+(-m)} = b^{0} = 1.

Derivation hint: use Step 1's result $b^0 = 1$ on the right. Therefore $b^{-m} = 1/b^{m}$ .

Step 3 — Fractional exponent. Apply [4] to $b^{1/n}$ raised to the $n$ th power:

9

\quad \left(b^{1/n}\right)^{n} = b^{(1/n)\cdot n} = b^{1} = b.

Derivation hint: multiply the exponents per [4], yielding exponent $1$ . So $b^{1/n}$ is the number whose $n$ th power is $b$ , i.e. $b^{1/n} = \sqrt[n]{b}$ . Combining with Step 2 extends this to all rationals: $b^{p/q} = \sqrt[q]{b^{p}}$ . Finally, irrational exponents are defined by continuity — taking limits of rational approximations — which is consistent because $b^x$ is monotone. $\blacksquare$

∎

mathematician · Leonhard Euler | 1707–1783

Euler systematized the modern theory of exponential and logarithmic functions in his 1748 Introductio in analysin infinitorum. He treated $a^x$ as a fully fledged function of a real variable, introduced and named the constant $e$ , and established the foundational identity $e^{ix} = \cos x + i \sin x$ , linking exponentials to trigonometry. His notation and viewpoint — exponents as continuous variables rather than mere counters — are essentially what we use today.

🧠 Intuition

Think of an exponential as repeated multiplication by the same factor, the multiplicative analogue of "add the same amount each step" (which gives a linear function). A linear function climbs by a fixed amount per step; an exponential climbs by a fixed percentage per step. That is why bank balances and bacteria look slow at first and then erupt: each step multiplies an already-larger number. The base $b$ tells the story — $b > 1$ grows, $0 < b < 1$ decays, and $b = 1$ stays flat. A useful mental anchor: with $b = 2$ the quantity doubles every unit of $x$ ; with $b = 1/2$ it halves.

Figure

Exponential growth (

y=2^x

, gold) versus linear growth (

y=2x

, grey). The curve starts below the line but rapidly overtakes it.

```svg

```

ExampleCompound interest — exponential growth

Given: $1000 invested at 5% annual interest, compounded yearly. Find the balance after 8 years, and the doubling time.

Setup. The balance multiplies by $1.05$ each year, so

10

\quad A(t) = 1000 \cdot (1.05)^{t}.

Derivation hint: 5% growth means the new amount is $100\% + 5\% = 105\% = 1.05$ of the old amount, applied $t$ times.

Step 1 — Evaluate at $t = 8$ :

A(8) = 1000 \cdot (1.05)^{8} = 1000 \cdot 1.477455 = 1477.46.

Step 2 — Doubling time solves $(1.05)^t = 2$ . Taking $\log$ :

t = \frac{\log 2}{\log 1.05} = \frac{0.30103}{0.021189} \approx 14.2.

\boxed{A(8) \approx \$1477.46;\quad \text{doubling time} \approx 14.2 \text{ years}}

ExampleRadioactive decay — exponential decay

Given: A sample of carbon-14 has half-life $5730$ years. Starting with $200$ g, how much remains after $10{,}000$ years?

Setup. Each half-life multiplies the amount by $\tfrac12$ , so with $t$ in years:

11

\quad N(t) = 200 \cdot \left(\tfrac{1}{2}\right)^{t/5730}.

Derivation hint: the exponent $t/5730$ counts how many half-lives have elapsed; each one halves the quantity.

Step 1 — Number of half-lives: $10000 / 5730 = 1.7452$ .

Step 2 — Evaluate:

N(10000) = 200 \cdot (0.5)^{1.7452} = 200 \cdot 0.2983 = 59.66.

\boxed{N(10000) \approx 59.7 \text{ g of } {}^{14}\mathrm{C}}

⚠️ Power functions are NOT exponential functions

A frequent error is confusing $x^n$ (a power function, variable base) with $b^x$ (an exponential function, variable exponent). They behave completely differently: $x^2$ grows polynomially, while $2^x$ eventually outgrows every polynomial. The laws of exponents apply to both notations, but you cannot, for instance, write $2^x = x^2$ or treat $(b^x)$ as if the base varied. Also beware: $b^{m+n} \neq b^m + b^n$ , and $(b+c)^n \neq b^n + c^n$ — exponents distribute over multiplication (Law [5]), never over addition.

CounterexampleThe addition fallacy

Suppose a student claims $(b^m)^n = b^{m+n}$ . Test with $b=2,\ m=3,\ n=2$ :

(2^3)^2 = 8^2 = 64, \qquad \text{but } 2^{3+2} = 2^5 = 32.

The correct law [4] gives $2^{3\cdot 2} = 2^6 = 64$ ✓. The exponents multiply when nesting powers; they only add when multiplying powers of the same base (Law [2]).

connection

Exponential functions are inseparable from their inverses, the Logarithmic Functions — taking $\log_b$ of $b^x$ returns $x$ , which is exactly how we solve equations like $b^x = c$ . The laws of exponents are mirror images of the Laws of Logarithms (multiplying becomes adding). The constant base $e$ and the function $e^x$ form the gateway to Differential Calculus, where $\tfrac{d}{dx}e^x = e^x$ . Discrete exponentials are precisely Geometric Sequences and Series, whose terms $a r^n$ are sampled values of $a b^x$ . They also generalize the Power Functions notation and underpin Compound Interest and the continuous-growth model.

🔬 Where this appears

⏳ History

- c. 250 BCE — Archimedes, in The Sand Reckoner, uses laws of exponents (adding indices of powers of ten) to express enormous numbers.

- 1484 — Nicolas Chuquet writes powers with explicit exponent notation, including a zero exponent, in Triparty.

- 1544 — Michael Stifel's Arithmetica integra tabulates powers of 2 alongside their exponents, foreshadowing logarithms.

- 1614 — John Napier publishes the first table of logarithms, turning multiplication of large numbers into addition.

- 1637 — René Descartes popularizes the modern superscript notation

x^3

in La Géométrie.

- 1655 — John Wallis interprets negative and fractional exponents systematically in Arithmetica infinitorum.

- 1683 — Jacob Bernoulli, studying continuous compound interest, encounters the limit that defines

e

.

- 1748 — Leonhard Euler, in Introductio in analysin infinitorum, names

e

, treats

a^x

as a continuous function, and unifies exponentials with trigonometry.

- 1873 — Charles Hermite proves

e

is transcendental, cementing the special status of the natural exponential base.

open-problem · Defining irrational and complex exponents rigorously

The "count the factors" picture breaks down once the exponent is irrational: what does $2^{\sqrt{2}}$ mean if you cannot multiply $2$ by itself " $\sqrt 2$ times"? The rigorous resolution requires limits — defining $b^x$ as the limit of $b^{q}$ over rationals $q \to x$ — or, more elegantly, defining $\exp(x) = \sum_{n=0}^\infty x^n/n!$ and $b^x = \exp(x \ln b)$ . A deeper subtlety: extending to complex exponents makes $b^z$ multivalued (since $\ln$ is multivalued), so $b^z$ requires choosing a branch. The Gelfond–Schneider theorem (1934) further shows that numbers like $2^{\sqrt 2}$ are transcendental, an area where many natural questions about exponentials remain delicate.

Exponential Functions & Laws of Exponents

Formal Statement

Why It Holds / Derivation

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also