Skip to content

Chapter 4 — Exponential and Logarithmic Functions

4.5 - Exponential and Logarithmic Equations

Solving an exponential or logarithmic equation requires no technique that this chapter has not already proved. It requires, instead, recognizing which of the tools already constructed corresponds to each situation, and not losing sight, along the way, of the domain.


The previous sections constructed two bijective functions, axa^x and its inverse log⁡ax\log_a x, and four laws that translate between the multiplicative world and the additive one. Solving an equation with these functions is, in essence, undoing layers until reaching an ordinary algebraic equation of the type already solved in §1.5. This section adds no new Propositions: it organizes, with worked examples, how to apply those already proved.

Principle of exponential injectivity. For a>0a>0, a≠1a\neq1, and functions f,gf,g with real values,

af(x)=ag(x)  ⟺  f(x)=g(x)a^{f(x)}=a^{g(x)} \iff f(x)=g(x)

Justification: the implication ⇐\Leftarrow is trivial, substituting equal values gives equal results; the implication ⇒\Rightarrow is exactly the injectivity of axa^x, §9 of section 4.1.

Principle of logarithmic injectivity. For a>0a>0, a≠1a\neq1, and functions f,gf,g with f(x)>0f(x)>0, g(x)>0g(x)>0,

log⁡af(x)=log⁡ag(x)  ⟺  f(x)=g(x)\log_a f(x)=\log_a g(x) \iff f(x)=g(x)

Justification: analogous, via the injectivity of log⁡a\log_a, §4 of section 4.3, with the caveat that here it is necessary to verify, before applying the principle, that both sides are in the domain of log⁡a\log_a, something that was not necessary for the exponential, defined on all of R\mathbb{R}.

Principle of direct application. If af(x)=ca^{f(x)}=c with c>0c>0, then f(x)=log⁡acf(x)=\log_a c, an immediate consequence of the Definition of §1, section 4.3: log⁡ac\log_a c is, by definition, the unique exponent that produces cc.


With the principles fixed, it is time to apply them to concrete equations, first exponential ones, distinguishing the case in which both sides reduce to the same base from the case in which they do not. It is the content of §2.

Example 1 (same base). Solve 53x−2=5x+45^{3x-2}=5^{x+4}.

Solution. By the Principle of exponential injectivity, 3x−2=x+43x-2=x+4, whence 2x=62x=6, that is, x=3x=3.

Example 2 (reducible to a common base). Solve 8x+1=162x−18^{x+1}=16^{2x-1}.

Solution. Both sides are rewritten in base 22: 8=238=2^3, 16=2416=2^4. By the power law between exponentials, (am)x=amx(a^m)^x=a^{mx}, corollary of §12, section 4.1, 8x+1=23(x+1)=23x+38^{x+1}=2^{3(x+1)}=2^{3x+3} and 162x−1=24(2x−1)=28x−416^{2x-1}=2^{4(2x-1)}=2^{8x-4}. The equation reduces to 23x+3=28x−42^{3x+3}=2^{8x-4}, and by injectivity, 3x+3=8x−43x+3=8x-4, whence 5x=75x=7, that is, x=75x=\dfrac75.

Example 3 (not reducible to a common base). Solve 72x+1=507^{2x+1}=50.

Solution. 5050 is not an integer or evident rational power of 77, so there is no common base to simplify. By the Principle of direct application, 2x+1=log⁡7502x+1=\log_7 50, whence

x=log⁡750−12x=\frac{\log_7 50-1}{2}

If a numerical value is desired, the change-of-base formula (§4, section 4.4) allows it to be computed with the decimal or natural logarithm: log⁡750=log⁡1050log⁡107\log_7 50=\dfrac{\log_{10}50}{\log_{10}7}.


Logarithmic equations require the same type of reasoning, but with a difference that cannot be omitted: each logarithm that appears in the statement imposes a domain restriction, and that restriction must be fixed before solving, not after. It is the content of §3.

Example 1. Solve log⁡3(x+5)=2\log_3(x+5)=2.

Solution. Domain: x+5>0x+5>0, that is, x>−5x>-5. By the Principle of direct application, read in the logarithmic direction: log⁡ay=k  ⟺  y=ak\log_a y=k \iff y=a^k, an immediate consequence of the Definition of §1, section 4.3, x+5=32=9x+5=3^2=9, whence x=4x=4. Since 4>−54>-5, the solution is valid.

Example 2. Solve log⁡2x+log⁡2(x−2)=3\log_2 x+\log_2(x-2)=3.

Solution. Domain: x>0x>0 and x−2>0x-2>0 simultaneously, that is, x>2x>2. By the product law (§1, section 4.4), the left side is log⁡2(x(x−2))\log_2\big(x(x-2)\big). The equation reduces to log⁡2(x2−2x)=3\log_2(x^2-2x)=3, and by the Principle of direct application, x2−2x=23=8x^2-2x=2^3=8, that is, x2−2x−8=0x^2-2x-8=0. Factoring, (x−4)(x+2)=0(x-4)(x+2)=0, whence x=4x=4 or x=−2x=-2. Only x=4x=4 satisfies the domain x>2x>2; x=−2x=-2 is discarded, not because the algebra is wrong, but because the product law, in combining two logarithms into one, cannot by itself remember the restriction that each imposed separately.


Some equations are not solved directly by the previous principles, but by recognizing, inside the equation, the structure of an already known algebraic equation, typically a quadratic, disguised by an exponential power. It is the content of §4.

Example. Solve e2x−3ex+2=0e^{2x}-3e^x+2=0.

Solution. Let t=ext=e^x. Since Rec⁡(ex)=(0,∞)\operatorname{Rec}(e^x)=(0,\infty), §10, section 4.1, with a=ea=e, the substitution requires t>0t>0. Since e2x=(ex)2=t2e^{2x}=(e^x)^2=t^2, the power law between exponentials, corollary of §12, section 4.1, the equation becomes

t2−3t+2=0t^2-3t+2=0

an ordinary quadratic equation of the type of §1.5. Factoring, (t−1)(t−2)=0(t-1)(t-2)=0, whence t=1t=1 or t=2t=2. Both values satisfy t>0t>0, so neither is discarded by the restriction of the substitution. Undoing the change: if t=1t=1, ex=1e^x=1, and by the Principle of direct application, or directly by log⁡e1=0\log_e1=0, §5 of section 4.3, x=0x=0. If t=2t=2, ex=2e^x=2, whence x=ln⁡2x=\ln2, using the usual notation ln⁡:=log⁡e\ln:=\log_e. The solutions are x=0x=0 and x=ln⁡2x=\ln2.


Exponential and logarithmic inequalities use the same principles of this section, with an additional ingredient: the direction of the inequality depends on the regime of the base, as already proved in §9 of section 4.1 and in §4 of section 4.3. It is the content of §5.

§5. Exponential and logarithmic inequalities

Section titled “§5. Exponential and logarithmic inequalities”

Example 1. Solve (13)x≤9\left(\dfrac13\right)^x\le9.

Solution. Since 9=32=(13)−29=3^2=\left(\dfrac13\right)^{-2}, the inequality is rewritten (13)x≤(13)−2\left(\dfrac13\right)^x\le\left(\dfrac13\right)^{-2}. Since 0<13<10<\dfrac13<1, the function is strictly decreasing, §9, section 4.1, and for a decreasing function, f(x)≤f(−2)f(x)\le f(-2) is equivalent to x≥−2x\ge-2, the same argument by contradiction as §4, section 4.3, applied here directly to the exponential instead of to the logarithm. The solution is x≥−2x\ge-2.

Example 2. Solve log⁡1/2x>−3\log_{1/2}x>-3, specifying the domain.

Solution. Domain: x>0x>0. Since −3=log⁡1/28-3=\log_{1/2}8, because (12)−3=8\left(\dfrac12\right)^{-3}=8, the inequality is rewritten log⁡1/2x>log⁡1/28\log_{1/2}x>\log_{1/2}8. Since 0<12<10<\dfrac12<1, log⁡1/2\log_{1/2} is strictly decreasing, §4, section 4.3, and for a decreasing function, f(x)>f(8)f(x)>f(8) is equivalent to x<8x<8. Combining with the domain, the solution is 0<x<80<x<8.


The entire section relied, without exception, on three facts already proved chapters, or sections, ago: that axa^x and log⁡ax\log_a x are mutually bijective (§4.1, §4.3), that this bijectivity translates equalities and inequalities between exponents into equalities and inequalities between powers, and vice versa (§9, section 4.1; §4, section 4.3), and that the operations translate into each other by means of the four laws of §4.4. There was not, in any example, a genuinely new technique, which, after four sections of rigorous construction, is exactly the expected reward: a well-constructed chapter makes solving, in the end, simpler than constructing.

What remains, for the section that closes the chapter, is the question of what all this is for outside mathematics itself: which phenomena of the world behave, genuinely, according to this structure of growth or decay proportional to the quantity itself, and not according to the additive accumulation that the polynomials of Chapter 3 described.


§4.6 closes the chapter with the applications of exponential and logarithmic functions: growth and decay, compound interest, and logarithmic scales.