Free CAT library

Laws of Indices, Power Towers & Exponential Equations

Master the fundamental algebraic laws of indices, fractional and negative exponents, the critical structural contrast between top-down power towers and nested exponentiation, exhaustive root-hunting across base parities, and auxiliary substitutions for non-linear exponential systems.

1. The Story & Mental Model: The Hyper-Scaling Engine of Repeated Multiplication

The Escalation from Addition to Hyper-Powers

Human intuition is fundamentally linear. When we walk, each step adds an equal distance: 1,2,3,4,5…1, 2, 3, 4, 5\dots This is the domain of addition.

When civilization required scaling across groups, mathematics invented multiplication—repeated addition:

k+k+k+⋯+k  (n times)=n⋅kk + k + k + \dots + k \; (n \text{ times}) = n \cdot k

Multiplication feels fast, but it is still constrained to constant multiplicative jumps. In the ancient Persian legend of the wheat and the chessboard, the inventor of chess asked the King for a seemingly modest reward: 11 grain of wheat on the first square, 22 on the second, 44 on the third, 88 on the fourth, doubling on each subsequent square up to the 64th64^{\text{th}}. The King laughed at the beggar's modesty—until the royal treasurers calculated the total:

1+2+4+⋯+263=264−1≈1.84×1019 grains of wheat1 + 2 + 4 + \dots + 2^{63} = 2^{64} - 1 \approx 1.84 \times 10^{19} \text{ grains of wheat}

This quantity represents roughly 1,2001,200 times the entire annual global wheat harvest today. The King was mathematically bankrupt.

This is the domain of indices (exponents): repeated multiplication.

an=a×a×a×⋯×a⏟n factorsa^n = \underbrace{a \times a \times a \times \dots \times a}_{n \text{ factors}}

The Physical Reality of Compounding Powers

Consider a standard sheet of paper approximately 0.1 mm0.1\text{ mm} thick.

  • Fold it in half 11 time: thickness is 0.2 mm0.2\text{ mm}.

  • Fold it 1010 times: thickness is 0.1×210=102.4 mm≈10.2 cm0.1 \times 2^{10} = 102.4\text{ mm} \approx 10.2\text{ cm}.

  • Fold it 2525 times: thickness exceeds 3.35 kilometers3.35\text{ kilometers} (higher than Mount Olympus).

  • Fold it 4242 times: the stack reaches beyond the Moon (384,000 km384,000\text{ km}).

  • Fold it 103103 times: the thickness exceeds the diameter of the observable universe (93 billion light years93\text{ billion light years}).

Linear arithmetic crawls; exponential indices teleport across orders of magnitude.

Why the Counter-Intuitive Rules Must Be True

Students often memorize index definitions by rote, but every rule is a logical necessity to keep arithmetic consistent:

  1. Why is a0=1a^0 = 1 (for a≠0a \ne 0)?

    Consider the quotient rule aman=am−n\frac{a^m}{a^n} = a^{m-n}. If we choose m=nm = n:

    amam=1andamam=am−m=a0\frac{a^m}{a^m} = 1 \quad \text{and} \quad \frac{a^m}{a^m} = a^{m-m} = a^0

    Therefore, to preserve the consistency of arithmetic division, a0a^0 must equal 11. (Note: 000^0 is indeterminate because 0/00/0 is undefined).

  2. Why is a−m=1ama^{-m} = \frac{1}{a^m}?

    a0am=1amanda0−m=a−m\frac{a^0}{a^m} = \frac{1}{a^m} \quad \text{and} \quad a^{0-m} = a^{-m}

    A negative exponent is not a negative number; it is an instruction to move in the reciprocal direction down the multiplicative ladder.

  3. Why is a1/n=ana^{1/n} = \sqrt[n]{a}?

    By the power of a power rule:

    (a1/n)n=a1n×n=a1=a\left(a^{1/n}\right)^n = a^{\frac{1}{n} \times n} = a^1 = a

    Thus, a1/na^{1/n} is by definition the unique positive number whose nthn^{\text{th}} power is aa. A fractional exponent partitions a single multiplicative leap into nn equal fractional multiplicative increments.


2. The CAT Question Blueprint: How Exam Setters Disguise Exponential Equations

In CAT and IPMAT, indices questions are rarely raw arithmetic calculations. Exam setters design questions around structural ambiguities and algebraic traps:

Blueprint 1: The Power Tower Mirage

Exam setters write power towers without brackets: abca^{b^c}.

  • Unwary students evaluate from left to right as (ab)c=abc(a^b)^c = a^{bc}.

  • The Reality: Exponentiation is strictly right-associative (top-down). That is:

    abc=a(bc)≠(ab)c=abca^{b^c} = a^{(b^c)} \ne (a^b)^c = a^{bc}

CAT questions exploit this by asking you to compare 2342^{3^4} and 2432^{4^3}, or to solve equations where nested brackets compete with top-down towers.

Blueprint 2: The Variable Base Parity Hunt (f(x)g(x)=1f(x)^{g(x)} = 1)

Questions of the form (f(x))g(x)=1(f(x))^{g(x)} = 1 appear frequently. An unprepared candidate will set g(x)=0g(x) = 0 and find 22 solutions. A 99th-percentile candidate knows that 11 can be produced in three distinct ways:

  1. The exponent is zero (g(x)=0g(x) = 0), provided the base is non-zero (f(x)≠0f(x) \ne 0).

  2. The base is one (f(x)=1f(x) = 1), for any real exponent g(x)g(x).

  3. The base is negative one (f(x)=−1f(x) = -1), provided the exponent g(x)g(x) evaluates to an even integer.

Failing to test f(x)=−1f(x) = -1 with even parity causes students to miss valid integer roots on test day!

Blueprint 3: The Parity Bifurcation in Equating Exponents (xn=ynx^n = y^n)

When equating xn=ynx^n = y^n:

  • If nn is odd, x=yx = y is the unique real solution.

  • If nn is even, x=±yx = \pm y (i.e., ∣x∣=∣y∣|x| = |y|).

Exam setters construct equations where discarding the negative branch eliminates half of the solution space.

Blueprint 4: Symmetric Exponential Substitution

When terms like axa^x and a−xa^{-x} (or 4x4^x and 4−x4^{-x}) appear in the same equation, exam setters disguise a quadratic equation behind exponential notation. Setting t=ax+a−xt = a^x + a^{-x} requires remembering that by the AM-GM inequality:

t=ax+1ax≥2(for ax>0)t = a^x + \frac{1}{a^x} \ge 2 \quad (\text{for } a^x > 0)

Any algebraic root of the quadratic yielding t<2t < 2 must be rejected!


3. Core Concepts, Formulae & Derivations

A. The Immutable Laws of Indices

For real bases a,b>0a, b > 0 and real exponents m,nm, n:

  1. Product of Identical Bases:

    am⋅an=am+na^m \cdot a^n = a^{m+n}

  2. Quotient of Identical Bases:

    aman=am−n(a≠0)\frac{a^m}{a^n} = a^{m-n} \quad (a \ne 0)

  3. Power of a Power:

    (am)n=amn=(an)m(a^m)^n = a^{mn} = (a^n)^m

  4. Power of a Product & Quotient:

    (ab)n=anbn,(ab)n=anbn(b≠0)(ab)^n = a^n b^n, \qquad \left(\frac{a}{b}\right)^n = \frac{a^n}{b^n} \quad (b \ne 0)

  5. Zero and Negative Exponents:

    a0=1(a≠0),a−m=1am(a≠0)a^0 = 1 \quad (a \ne 0), \qquad a^{-m} = \frac{1}{a^m} \quad (a \ne 0)

  6. Fractional Exponents (Radicals):

    ap/q=apq=(aq)p(a>0,q∈Z+)a^{p/q} = \sqrt[q]{a^p} = \left(\sqrt[q]{a}\right)^p \quad (a > 0, q \in \mathbb{Z}^+)

B. Power Towers vs Nested Exponents

There is a profound difference between nested parentheses and chained exponents:

  • Nested Expression: ((a)b)c=ab⋅c((a)^b)^c = a^{b \cdot c}. Here, exponents multiply.

  • Power Tower: abc=a(bc)a^{b^c} = a^{\left(b^c\right)}. Exponentiation is evaluated from the top down.

Numerical Proof of Disparity:

Let a=2,b=3,c=2a = 2, b = 3, c = 2:

Nested: (23)2=82=64=23×2=26\text{Nested: } (2^3)^2 = 8^2 = 64 = 2^{3 \times 2} = 2^6

Power Tower: 232=2(9)=512\text{Power Tower: } 2^{3^2} = 2^{(9)} = 512

Notice that 512=8×64512 = 8 \times 64. The power tower is 88 times larger! For higher exponents, the difference becomes astronomical.

C. Exhaustive Solution Cases for (f(x))g(x)=1(f(x))^{g(x)} = 1

To find all real solutions to (f(x))g(x)=1(f(x))^{g(x)} = 1, test three mutually exclusive branches:

\mathbf{Branch\; 1:} & g(x) = 0 \quad \text{and} \quad f(x) \ne 0 \\[4pt] \mathbf{Branch\; 2:} & f(x) = 1 \quad \text{for any real } g(x) \\[4pt] \mathbf{Branch\; 3:} & f(x) = -1 \quad \text{and } g(x) \in 2\mathbb{Z} \; (g(x) \text{ is an even integer}) \end{cases}$$ #### D. Exhaustive Solution Cases for $(f(x))^{g(x)} = (f(x))^{h(x)}$ When bases are identical expressions containing the variable: $$\begin{cases} \mathbf{Branch\; 1:} & g(x) = h(x) \quad (\text{provided } f(x) \ne 0) \\[4pt] \mathbf{Branch\; 2:} & f(x) = 1 \\[4pt] \mathbf{Branch\; 3:} & f(x) = -1 \quad (\text{provided } g(x) \text{ and } h(x) \text{ are integers of identical parity}) \\[4pt] \mathbf{Branch\; 4:} & f(x) = 0 \quad (\text{provided } g(x) > 0 \text{ and } h(x) > 0) \end{cases}$$ #### E. Decoupling Systems of Exponential Equations For simultaneous systems where variables are intertwined across bases, such as: $$a^x b^y = K_1 \quad \text{and} \quad a^y b^x = K_2$$ **The Decoupling Strategy:** 1. **Multiply** the two equations: $$(a^x b^y)(a^y b^x) = (ab)^{x+y} = K_1 K_2$$ 2. **Divide** the two equations: $$\frac{a^x b^y}{a^y b^x} = \left(\frac{a}{b}\right)^{x-y} = \frac{K_1}{K_2}$$ This immediately decouples the non-linear system into two elementary linear equations in $(x+y)$ and $(x-y)$. --- ### 4. Common Pitfalls & Speed Shortcuts #### Pitfall 1: Distributing Exponents Over Addition - **The Blunder:** Writing $(x + y)^n = x^n + y^n$. - **The Reality:** Powers distribute **only over multiplication and division**, never over addition or subtraction. For example, $(3 + 4)^2 = 7^2 = 49$, whereas $3^2 + 4^2 = 9 + 16 = 25$. #### Pitfall 2: Forgetting the $t \ge 2$ Bound on $a^x + a^{-x}$ - When solving an equation by setting $t = 2^x + 2^{-x}$, candidates solve a quadratic $t^2 - 3t + 2 = 0 \implies t = 1 \text{ or } 2$. - They then attempt to solve $2^x + 2^{-x} = 1$. But by the AM-GM inequality, for any positive real number $y = 2^x$: $$\frac{y + \frac{1}{y}}{2} \ge \sqrt{y \cdot \frac{1}{y}} = 1 \implies y + \frac{1}{y} \ge 2$$ - The minimum possible value of $2^x + 2^{-x}$ is $2$ (achieved at $x = 0$). Thus $t = 1$ has **no real solution**! Blindly accepting both roots leads to incorrect answer choices. #### Speed Shortcut 1: The Exponent GCD Technique for Giant Powers When asked to compare gigantic powers like $3^{75}$, $2^{125}$, $5^{50}$: - Do not attempt logarithms unless necessary. - Find the Greatest Common Divisor of the exponents: $\gcd(75, 125, 50) = 25$. - Rewrite each power with the common exponent $25$: $$3^{75} = (3^3)^{25} = 27^{25}, \quad 2^{125} = (2^5)^{25} = 32^{25}, \quad 5^{50} = (5^2)^{25} = 25^{25}$$ - Since all exponents are now identical and positive, simply order the bases: $25 < 27 < 32 \implies 5^{50} < 3^{75} < 2^{125}$. The entire comparison takes less than $20$ seconds! #### Speed Shortcut 2: Base Homogenization via Prime Factorization Always express every numerical base in terms of its canonical prime factors: $4 = 2^2$, $6 = 2 \cdot 3$, $9 = 3^2$, $12 = 2^2 \cdot 3$, $18 = 2 \cdot 3^2$. Equating exponents of like primes on both sides reduces any exponential maze into a predictable system of linear equations.

Formulas

Fundamental Laws of Indices

am⋅an=am+n,aman=am−n,(am)n=amna^m \cdot a^n = a^{m+n}, \quad \frac{a^m}{a^n} = a^{m-n}, \quad (a^m)^n = a^{mn}

Valid for all positive real bases a > 0
Power Tower (Right-Associative)

abc=a(bc)≠(ab)c=abca^{b^c} = a^{\left(b^c\right)} \ne (a^b)^c = a^{bc}

Evaluated strictly top-to-bottom unless parentheses indicate otherwise
Three-Branch Rule for Base-Exponent Unity

(f(x))g(x)=1  ⟺  {g(x)=0  (f(x)≠0)f(x)=1f(x)=−1  (g(x)∈2Z)(f(x))^{g(x)} = 1 \iff \begin{cases} g(x) = 0 \; (f(x) \ne 0) \\ f(x) = 1 \\ f(x) = -1 \; (g(x) \in 2\mathbb{Z}) \end{cases}

Always check whether g(x) is an even integer when f(x) = -1
Symmetric Exponential Substitution

t=ax+a−x  ⟹  a2x+a−2x=t2−2,with t≥2t = a^x + a^{-x} \implies a^{2x} + a^{-2x} = t^2 - 2, \quad \text{with } t \ge 2

AM-GM guarantees t >= 2 for positive real a^x; discard roots with t < 2
Decoupling Coupled Systems

(axby)(aybx)=(ab)x+y,axbyaybx=(ab)x−y(a^x b^y)(a^y b^x) = (ab)^{x+y}, \quad \frac{a^x b^y}{a^y b^x} = \left(\frac{a}{b}\right)^{x-y}

Multiplication and division convert coupled exponential systems into linear equations

Traps and tips

  • Right-Associativity Rule: Always compute power towers from the top down: a^{b^c} = a^{(b^c)}, never (a^b)^c.
  • The Three Branches of 1: When solving f(x)^{g(x)} = 1, test: (1) g(x) = 0 with f(x) != 0, (2) f(x) = 1, and (3) f(x) = -1 with g(x) even.
  • AM-GM Bound on Symmetric Terms: Setting t = a^x + a^{-x} strictly requires t >= 2. Any quadratic solution giving t < 2 yields zero real values of x.
  • Exponent GCD Trick: Compare massive powers by dividing all exponents by their GCD to normalize them to a shared exponent level.
  • Prime Factorization First: When bases differ (e.g. 12, 18, 24), immediately decompose them into primes (2^p * 3^q) to equate independent exponents.

Solved examples

Example 1.1: Power Towers vs Nested Powers & Tower Equations

(a) Evaluate and compare the values of P=232P = 2^{3^2} and Q=(23)2Q = (2^3)^2.

(b) Find all positive integer solutions nn to the power-tower equation: (22)n=22n(2^2)^n = 2^{2^n}.

Show solution

Part (a): Evaluating PP and QQ

Step 1: Evaluate P=232P = 2^{3^2}.

Because exponentiation is right-associative (evaluated top-down), the exponent tower 323^2 must be calculated first:

32=9  ⟹  P=29=5123^2 = 9 \implies P = 2^9 = 512

Step 2: Evaluate Q=(23)2Q = (2^3)^2.

Here, parentheses dictate that the base 232^3 is computed first, or we apply the power of a power law (am)n=amn(a^m)^n = a^{mn}:

(23)2=23×2=26=64(2^3)^2 = 2^{3 \times 2} = 2^6 = 64

Step 3: Compare PP and QQ.

PQ=51264=8\frac{P}{Q} = \frac{512}{64} = 8

The power tower 2322^{3^2} is 88 times larger than (23)2(2^3)^2.


Part (b): Solving (22)n=22n(2^2)^n = 2^{2^n} for n∈Z+n \in \mathbb{Z}^+

Step 1: Simplify the Left-Hand Side.

Using (am)n=amn(a^m)^n = a^{mn}:

LHS=(22)n=22n\text{LHS} = (2^2)^n = 2^{2n}

Step 2: Express the Right-Hand Side.

RHS=22n=2(2n)\text{RHS} = 2^{2^n} = 2^{\left(2^n\right)}

Step 3: Equate the exponents of the common base 22.

Since the base is 2>12 > 1, we equate the exponents directly:

2n=2n2n = 2^n

Step 4: Test positive integer candidates n∈{1,2,3,… }n \in \{1, 2, 3, \dots\}.

  • For n=1n = 1: 2(1)=22(1) = 2 and 21=22^1 = 2. Both sides equal 22. (Valid)

  • For n=2n = 2: 2(2)=42(2) = 4 and 22=42^2 = 4. Both sides equal 44. (Valid)

  • For n=3n = 3: 2(3)=62(3) = 6, while 23=82^3 = 8. Here 2n>2n2^n > 2n.

  • For n≥3n \ge 3: By mathematical induction, 2n>2n2^n > 2n for all integers n≥3n \ge 3 because 2n2^n doubles with each step while 2n2n increases by only 22.

Thus, the only positive integer solutions are n=1n = 1 and n=2n = 2.

Takeaway: Exponentiation without parentheses is strictly right-associative (top-down): a^b^c = a^(b^c) , which scales exponentially faster than (a^b)^c = a^bc .

Example 1.2: Variable Base & Variable Exponent Equations

Find the total number of distinct real solutions to the equation:

(x2−7x+11)x2−11x+30=1(x^2 - 7x + 11)^{x^2 - 11x + 30} = 1

Show solution

Let the base be f(x)=x2−7x+11f(x) = x^2 - 7x + 11 and the exponent be g(x)=x2−11x+30g(x) = x^2 - 11x + 30.

An equation of the form (f(x))g(x)=1(f(x))^{g(x)} = 1 has solutions arising from three mutually exclusive branches.


Branch 1: The Exponent is Zero (g(x)=0g(x) = 0), provided f(x)≠0f(x) \ne 0.

x2−11x+30=0x^2 - 11x + 30 = 0

Factorizing the quadratic:

(x−5)(x−6)=0  ⟹  x=5orx=6(x - 5)(x - 6) = 0 \implies x = 5 \quad \text{or} \quad x = 6

Now verify that the base f(x)≠0f(x) \ne 0 for each candidate:

  • For x=5x = 5: f(5)=52−7(5)+11=25−35+11=1≠0f(5) = 5^2 - 7(5) + 11 = 25 - 35 + 11 = 1 \ne 0. (Valid root: x=5x = 5)

  • For x=6x = 6: f(6)=62−7(6)+11=36−42+11=5≠0f(6) = 6^2 - 7(6) + 11 = 36 - 42 + 11 = 5 \ne 0. (Valid root: x=6x = 6)


Branch 2: The Base is Equal to 11 (f(x)=1f(x) = 1).

x2−7x+11=1  ⟹  x2−7x+10=0x^2 - 7x + 11 = 1 \implies x^2 - 7x + 10 = 0

Factorizing the quadratic:

(x−2)(x−5)=0  ⟹  x=2orx=5(x - 2)(x - 5) = 0 \implies x = 2 \quad \text{or} \quad x = 5

Since 1anything=11^{\text{anything}} = 1, both are valid. (x=5x = 5 was already found in Branch 1; x=2x = 2 is new).


Branch 3: The Base is Equal to −1-1 (f(x)=−1f(x) = -1), provided g(x)g(x) is an even integer.

x2−7x+11=−1  ⟹  x2−7x+12=0x^2 - 7x + 11 = -1 \implies x^2 - 7x + 12 = 0

Factorizing the quadratic:

(x−3)(x−4)=0  ⟹  x=3orx=4(x - 3)(x - 4) = 0 \implies x = 3 \quad \text{or} \quad x = 4

Now verify the parity of the exponent g(x)=x2−11x+30g(x) = x^2 - 11x + 30 for each candidate:

  • For x=3x = 3:

    g(3)=32−11(3)+30=9−33+30=6g(3) = 3^2 - 11(3) + 30 = 9 - 33 + 30 = 6

    Since 66 is an even integer, (−1)6=1(-1)^6 = 1. (Valid root: x=3x = 3)

  • For x=4x = 4:

    g(4)=42−11(4)+30=16−44+30=2g(4) = 4^2 - 11(4) + 30 = 16 - 44 + 30 = 2

    Since 22 is an even integer, (−1)2=1(-1)^2 = 1. (Valid root: x=4x = 4)


Step 4: Consolidate the Distinct Real Solutions.

Collecting all unique roots across the three branches:

S={2,3,4,5,6}\mathcal{S} = \{2, 3, 4, 5, 6\}

The equation has 55 distinct real solutions.

Takeaway: Always examine all 3 exhaustive branches for f(x)^g(x) = 1 : exponent zero (base 0 ), base equals 1 , and base equals -1 (exponent must evaluate to an even integer).

Example 1.3: Quadratic Reduction via Exponential Substitution

Find all real solutions xx satisfying the exponential equation:

4x+0.5−7⋅2x−4=04^{x + 0.5} - 7 \cdot 2^x - 4 = 0

Show solution

Step 1: Decompose the exponential terms into powers of 2x2^x.

Notice that 4=224 = 2^2. Therefore:

4x+0.5=4x⋅40.5=(22x)⋅(4)=2⋅(2x)24^{x + 0.5} = 4^x \cdot 4^{0.5} = (2^{2x}) \cdot (\sqrt{4}) = 2 \cdot (2^x)^2

Step 2: Formulate the auxiliary variable substitution.

Let u=2xu = 2^x.

Crucial Domain Constraint: Because 2x>02^x > 0 for all real xx, we must have u>0u > 0.

Substituting uu into the equation:

2u2−7u−4=02u^2 - 7u - 4 = 0

Step 3: Solve the quadratic equation in uu.

Factor by splitting the middle term (2×(−4)=−82 \times (-4) = -8, factors −8-8 and +1+1):

2u2−8u+u−4=02u^2 - 8u + u - 4 = 0

2u(u−4)+1(u−4)=02u(u - 4) + 1(u - 4) = 0

(2u+1)(u−4)=0(2u + 1)(u - 4) = 0

This yields two algebraic roots:

u=4oru=−12u = 4 \quad \text{or} \quad u = -\frac{1}{2}

Step 4: Apply the positivity domain filter and solve for xx.

  • Case 1: u=−12u = -\frac{1}{2}

    Since 2x>02^x > 0 for all real xx, 2x=−122^x = -\frac{1}{2} has no real solution. (Extraneous root discarded).

  • Case 2: u=4u = 4

    2x=4=22  ⟹  x=22^x = 4 = 2^2 \implies x = 2

Verification:

Substitute x=2x = 2 back into the original equation:

42+0.5−7⋅22−4=42.5−7(4)−4=32−28−4=0(Verified)4^{2 + 0.5} - 7 \cdot 2^2 - 4 = 4^{2.5} - 7(4) - 4 = 32 - 28 - 4 = 0 \quad \text{(Verified)}

The unique real solution is x=2x = 2.

Takeaway: When substituting u = a^x , strictly enforce u 0 ; negative algebraic roots have no real inverse and must be discarded immediately.

Example 1.4: Symmetric Reciprocal Exponentials & AM-GM Constraints

Solve for all real values of xx:

2(4x+4−x)−9(2x+2−x)+14=02(4^x + 4^{-x}) - 9(2^x + 2^{-x}) + 14 = 0

Show solution

Step 1: Recognize the reciprocal symmetry and define the auxiliary variable.

Let t=2x+2−xt = 2^x + 2^{-x}.

By the AM-GM inequality, for any positive quantity y=2x>0y = 2^x > 0:

y+1y2≥y⋅1y=1  ⟹  y+1y≥2\frac{y + \frac{1}{y}}{2} \ge \sqrt{y \cdot \frac{1}{y}} = 1 \implies y + \frac{1}{y} \ge 2

Hence, any real value of xx strictly requires t≥2t \ge 2.

Step 2: Express 4x+4−x4^x + 4^{-x} in terms of tt.

Square both sides of t=2x+2−xt = 2^x + 2^{-x}:

t2=(2x+2−x)2=(2x)2+2(2x)(2−x)+(2−x)2=4x+2+4−xt^2 = (2^x + 2^{-x})^2 = (2^x)^2 + 2(2^x)(2^{-x}) + (2^{-x})^2 = 4^x + 2 + 4^{-x}

4x+4−x=t2−24^x + 4^{-x} = t^2 - 2

Step 3: Substitute into the original equation and solve for tt.

2(t2−2)−9t+14=02(t^2 - 2) - 9t + 14 = 0

2t2−4−9t+14=02t^2 - 4 - 9t + 14 = 0

2t2−9t+10=02t^2 - 9t + 10 = 0

Factorizing the quadratic (2×10=202 \times 10 = 20, factors −4-4 and −5-5):

2t2−4t−5t+10=02t^2 - 4t - 5t + 10 = 0

2t(t−2)−5(t−2)=02t(t - 2) - 5(t - 2) = 0

(2t−5)(t−2)=0  ⟹  t=2ort=52=2.5(2t - 5)(t - 2) = 0 \implies t = 2 \quad \text{or} \quad t = \frac{5}{2} = 2.5

Both values satisfy the AM-GM feasibility condition t≥2t \ge 2!


Step 4: Back-substitute to find xx.

Case A: t=2t = 2

2x+2−x=22^x + 2^{-x} = 2

Multiplying by 2x2^x (or recognizing the AM-GM equality condition):

(2x)2−2(2x)+1=0  ⟹  (2x−1)2=0  ⟹  2x=1  ⟹  x=0(2^x)^2 - 2(2^x) + 1 = 0 \implies (2^x - 1)^2 = 0 \implies 2^x = 1 \implies x = 0

Case B: t=52t = \frac{5}{2}

2x+12x=522^x + \frac{1}{2^x} = \frac{5}{2}

Let u=2xu = 2^x:

u+1u=52  ⟹  2u2−5u+2=0u + \frac{1}{u} = \frac{5}{2} \implies 2u^2 - 5u + 2 = 0

(2u−1)(u−2)=0  ⟹  u=2oru=12(2u - 1)(u - 2) = 0 \implies u = 2 \quad \text{or} \quad u = \frac{1}{2}

  • If u=2  ⟹  2x=21  ⟹  x=1u = 2 \implies 2^x = 2^1 \implies x = 1

  • If u=12  ⟹  2x=2−1  ⟹  x=−1u = \frac{1}{2} \implies 2^x = 2^{-1} \implies x = -1


Step 5: Consolidate the Solution Set.

S={−1,0,1}\mathcal{S} = \{-1, 0, 1\}

The complete set of real solutions is x∈{−1,0,1}x \in \{-1, 0, 1\}.

Takeaway: Substitute t = a^x + a^-x with a^2x + a^-2x = t^2 - 2 ; always enforce t 2 from AM-GM before back-solving for real x .

Example 1.5: Cross-Coupled Exponential Systems

Find all real pairs (x,y)(x, y) that satisfy the simultaneous system of exponential equations:

2x⋅3y=182^x \cdot 3^y = 18

2y⋅3x=122^y \cdot 3^x = 12

Show solution

Step 1: Understand the symmetry of the coupled system.

We are given:

(1) 2x⋅3y=18=21⋅322^x \cdot 3^y = 18 = 2^1 \cdot 3^2

(2) 2y⋅3x=12=22⋅312^y \cdot 3^x = 12 = 2^2 \cdot 3^1

The variables xx and yy are symmetrically coupled across the prime bases 22 and 33.


Step 2: Multiply equation (1) and equation (2) to isolate (x+y)(x + y).

(2x⋅3y)⋅(2y⋅3x)=18×12(2^x \cdot 3^y) \cdot (2^y \cdot 3^x) = 18 \times 12

Group by like bases:

(2x+y)⋅(3x+y)=216(2^{x+y}) \cdot (3^{x+y}) = 216

Combine under a single base (2×3)(2 \times 3):

6x+y=2166^{x+y} = 216

Since 216=63216 = 6^3:

6x+y=63  ⟹  x+y=3— (Equation 3)6^{x+y} = 6^3 \implies x + y = 3 \quad \text{--- (Equation 3)}


Step 3: Divide equation (1) by equation (2) to isolate (x−y)(x - y).

2x⋅3y2y⋅3x=1812=32\frac{2^x \cdot 3^y}{2^y \cdot 3^x} = \frac{18}{12} = \frac{3}{2}

Rewrite the LHS by collecting like terms:

2x−y⋅3y−x=322^{x-y} \cdot 3^{y-x} = \frac{3}{2}

Notice that 3y−x=13x−y=3−(x−y)3^{y-x} = \frac{1}{3^{x-y}} = 3^{-(x-y)}. Thus:

2x−y3x−y=(23)x−y=32\frac{2^{x-y}}{3^{x-y}} = \left(\frac{2}{3}\right)^{x-y} = \frac{3}{2}

Express the RHS with base 23\frac{2}{3}:

32=(23)−1\frac{3}{2} = \left(\frac{2}{3}\right)^{-1}

Therefore:

(23)x−y=(23)−1  ⟹  x−y=−1— (Equation 4)\left(\frac{2}{3}\right)^{x-y} = \left(\frac{2}{3}\right)^{-1} \implies x - y = -1 \quad \text{--- (Equation 4)}


Step 4: Solve the resulting linear system.

Add Equation (3) and Equation (4):

(x+y)+(x−y)=3+(−1)  ⟹  2x=2  ⟹  x=1(x + y) + (x - y) = 3 + (-1) \implies 2x = 2 \implies x = 1

Substitute x=1x = 1 into Equation (3):

1+y=3  ⟹  y=21 + y = 3 \implies y = 2

Step 5: Verify in the original equations.

  • Eq (1): 21⋅32=2⋅9=182^1 \cdot 3^2 = 2 \cdot 9 = 18 (Correct)

  • Eq (2): 22⋅31=4⋅3=122^2 \cdot 3^1 = 4 \cdot 3 = 12 (Correct)

The unique real solution pair is (x,y)=(1,2)(x, y) = (1, 2).

Takeaway: For cross-coupled exponential systems a^x b^y = K 1 and a^y b^x = K 2 , multiply and divide the equations to immediately decouple into linear equations in (x+y) and (x-y) .

Example 1.6: Exponential Magnitude Ordering & Polynomial Inequalities

(a) Arrange the following quantities in ascending order: A=375A = 3^{75}, B=2125B = 2^{125}, C=550C = 5^{50}, D=1025D = 10^{25}.

(b) Find the complete set of integer solutions xx satisfying the inequality: 32x+1−28⋅3x+9≤03^{2x+1} - 28 \cdot 3^x + 9 \le 0.

Show solution

Part (a): Ordering Astronomical Powers

Step 1: Identify the GCD of the exponents.

The exponents are 75,125,50,2575, 125, 50, 25.

gcd⁡(75,125,50,25)=25\gcd(75, 125, 50, 25) = 25

Step 2: Rewrite each quantity with the common exponent 2525.

  • A=375=(33)25=2725A = 3^{75} = (3^3)^{25} = 27^{25}

  • B=2125=(25)25=3225B = 2^{125} = (2^5)^{25} = 32^{25}

  • C=550=(52)25=2525C = 5^{50} = (5^2)^{25} = 25^{25}

  • D=1025=(101)25=1025D = 10^{25} = (10^1)^{25} = 10^{25}

Step 3: Compare the bases directly.

Since the exponent 25>025 > 0 is uniform across all four quantities, the order of the numbers is determined strictly by their bases:

10<25<27<3210 < 25 < 27 < 32

Therefore:

1025<2525<2725<3225  ⟹  D<C<A<B10^{25} < 25^{25} < 27^{25} < 32^{25} \implies D < C < A < B


Part (b): Solving 32x+1−28⋅3x+9≤03^{2x+1} - 28 \cdot 3^x + 9 \le 0 for x∈Zx \in \mathbb{Z}

Step 1: Transform into an auxiliary polynomial in u=3xu = 3^x.

Using index laws: 32x+1=31⋅(3x)2=3(3x)23^{2x+1} = 3^1 \cdot (3^x)^2 = 3(3^x)^2.

Let u=3x>0u = 3^x > 0. The inequality becomes:

3u2−28u+9≤03u^2 - 28u + 9 \le 0

Step 2: Factorize the quadratic polynomial.

Product =3×9=27= 3 \times 9 = 27, Sum =−28= -28. The factors are −27-27 and −1-1:

3u2−27u−u+9≤03u^2 - 27u - u + 9 \le 0

3u(u−9)−1(u−9)≤03u(u - 9) - 1(u - 9) \le 0

(3u−1)(u−9)≤0(3u - 1)(u - 9) \le 0

Step 3: Determine the feasible interval for uu.

13≤u≤9\frac{1}{3} \le u \le 9

Step 4: Back-substitute u=3xu = 3^x and solve for xx.

Express the boundary values as powers of 33:

3−1≤3x≤323^{-1} \le 3^x \le 3^2

Since the base 3>13 > 1, the exponential function is strictly increasing, preserving inequality signs:

−1≤x≤2-1 \le x \le 2

Step 5: Extract the integer solutions.

x∈{−1,0,1,2}x \in \{-1, 0, 1, 2\}

There are 44 integer solutions: x∈{−1,0,1,2}x \in \{-1, 0, 1, 2\}.

Takeaway: To compare massive powers, divide exponents by their GCD to equalize the exponent layer; for exponential inequalities, solve the reduced polynomial first, then invert monotonically.

Practice questions

Question 1 · CAT Level

Find the sum of all distinct real solutions to the equation:

(x2−5x+5)x2−9x+20=1(x^2 - 5x + 5)^{x^2 - 9x + 20} = 1

  1. A

    9

  2. B

    10

  3. C

    14

  4. D

    15

Show answer and explanation

Answer: (D) 15

An equation of the form (f(x))g(x)=1(f(x))^{g(x)} = 1 generates solutions from three exhaustive branches.

Branch 1: Exponent g(x)=0  ⟹  x2−9x+20=0  ⟹  (x−4)(x−5)=0g(x) = 0 \implies x^2 - 9x + 20 = 0 \implies (x - 4)(x - 5) = 0, giving x=4x = 4 and x=5x = 5. At both values, base f(x)≠0f(x) \ne 0.

Branch 2: Base f(x)=1  ⟹  x2−5x+5=1  ⟹  x2−5x+4=0  ⟹  (x−1)(x−4)=0f(x) = 1 \implies x^2 - 5x + 5 = 1 \implies x^2 - 5x + 4 = 0 \implies (x - 1)(x - 4) = 0, yielding x=1x = 1 and x=4x = 4.

Branch 3: Base f(x)=−1  ⟹  x2−5x+5=−1  ⟹  x2−5x+6=0  ⟹  (x−2)(x−3)=0f(x) = -1 \implies x^2 - 5x + 5 = -1 \implies x^2 - 5x + 6 = 0 \implies (x - 2)(x - 3) = 0. Testing exponent parity: g(2)=6g(2) = 6 (even) and g(3)=2g(3) = 2 (even), so both x=2x = 2 and x=3x = 3 are valid.

Collecting all distinct real solutions across the branches gives {1,2,3,4,5}\{1, 2, 3, 4, 5\}, whose sum is 1+2+3+4+5=151 + 2 + 3 + 4 + 5 = 15.

Question 2 · CAT Level

Find the sum of all positive real solutions to the power-tower equation:

x(xx)=(xx)xx^{\left(x^{\sqrt{x}}\right)} = \left(x^{\sqrt{x}}\right)^x

  1. A

    94\frac{9}{4}

  2. B

    134\frac{13}{4}

  3. C

    174\frac{17}{4}

  4. D

    254\frac{25}{4}

Show answer and explanation

Answer: (B) 13/4

The LHS is a top-down power tower x(xx)x^{\left(x^{\sqrt{x}}\right)}, while the RHS simplifies via nested power laws: (xx)x=xx⋅x=xx3/2\left(x^{\sqrt{x}}\right)^x = x^{\sqrt{x} \cdot x} = x^{x^{3/2}}.

First, consider the base root: for x=1x = 1, 111=(11)1=11^{1^1} = (1^1)^1 = 1, so x=1x = 1 is a valid positive real solution.

For x>0x > 0 with x≠1x \ne 1, equating the exponents gives xx=x3/2x^{\sqrt{x}} = x^{3/2}.

Again equating the exponents of base xx yields x=32  ⟹  x=(32)2=94\sqrt{x} = \frac{3}{2} \implies x = \left(\frac{3}{2}\right)^2 = \frac{9}{4}.

The complete set of positive real solutions is {1,94}\left\{1, \frac{9}{4}\right\}, and their sum is 1+94=1341 + \frac{9}{4} = \frac{13}{4}.

Question 3 · CAT Level

If xx and yy are real numbers satisfying the simultaneous exponential system:

2x⋅3y=1622^x \cdot 3^y = 162

2y⋅3x=482^y \cdot 3^x = 48

find the value of x2+y2x^2 + y^2.

  1. A

    13

  2. B

    17

  3. C

    20

  4. D

    25

Show answer and explanation

Answer: (B) 17

Multiplying the two equations gives (2x⋅3y)(2y⋅3x)=162×48=7776(2^x \cdot 3^y)(2^y \cdot 3^x) = 162 \times 48 = 7776.

Grouping like bases yields (2⋅3)x+y=6x+y=7776=65  ⟹  x+y=5(2 \cdot 3)^{x+y} = 6^{x+y} = 7776 = 6^5 \implies x + y = 5.

Dividing the first equation by the second gives 2x⋅3y2y⋅3x=(23)x−y=16248=278=(23)−3  ⟹  x−y=−3\frac{2^x \cdot 3^y}{2^y \cdot 3^x} = \left(\frac{2}{3}\right)^{x-y} = \frac{162}{48} = \frac{27}{8} = \left(\frac{2}{3}\right)^{-3} \implies x - y = -3.

Adding and subtracting the decoupled linear equations gives 2x=2  ⟹  x=12x = 2 \implies x = 1 and 2y=8  ⟹  y=42y = 8 \implies y = 4.

Thus, the value of x2+y2=12+42=1+16=17x^2 + y^2 = 1^2 + 4^2 = 1 + 16 = 17.

Question 4 · CAT Level

Find the product of all real solutions to the equation:

(5+26)x2−3+(5−26)x2−3=10(5 + 2\sqrt{6})^{x^2 - 3} + (5 - 2\sqrt{6})^{x^2 - 3} = 10

  1. A

    4

  2. B

    6

  3. C

    8

  4. D

    16

Show answer and explanation

Answer: (C) 8

Notice that (5+26)(5−26)=25−24=1(5 + 2\sqrt{6})(5 - 2\sqrt{6}) = 25 - 24 = 1, meaning 5−26=(5+26)−15 - 2\sqrt{6} = (5 + 2\sqrt{6})^{-1}.

Letting u=(5+26)x2−3u = (5 + 2\sqrt{6})^{x^2 - 3}, the equation reduces to u+1u=10  ⟹  u2−10u+1=0u + \frac{1}{u} = 10 \implies u^2 - 10u + 1 = 0.

Solving the quadratic gives u=5±24=5±26=(5+26)±1u = 5 \pm \sqrt{24} = 5 \pm 2\sqrt{6} = (5 + 2\sqrt{6})^{\pm 1}.

Equating exponents: x2−3=1  ⟹  x2=4  ⟹  x=±2x^2 - 3 = 1 \implies x^2 = 4 \implies x = \pm 2, and x2−3=−1  ⟹  x2=2  ⟹  x=±2x^2 - 3 = -1 \implies x^2 = 2 \implies x = \pm\sqrt{2}.

All four roots {−2,2,−2,2}\{-2, 2, -\sqrt{2}, \sqrt{2}\} are real, and their product is (−2)×2×(−2)×2=(−4)×(−2)=8(-2) \times 2 \times (-\sqrt{2}) \times \sqrt{2} = (-4) \times (-2) = 8.

Question 5 · CAT Level

Find the sum of all integer solutions to the inequality:

(x2−4x+3)⋅2x2−5x+6≤x2−4x+3(x^2 - 4x + 3) \cdot 2^{x^2 - 5x + 6} \le x^2 - 4x + 3

  1. A

    3

  2. B

    5

  3. C

    6

  4. D

    7

Show answer and explanation

Answer: (C) 6

Rearranging terms to the left-hand side yields (x2−4x+3)(2x2−5x+6−1)≤0(x^2 - 4x + 3)\left(2^{x^2 - 5x + 6} - 1\right) \le 0.

Since 2u−12^u - 1 shares the exact same sign as uu for all real uu, the sign of the exponential factor matches x2−5x+6x^2 - 5x + 6.

Factoring both quadratics transforms the inequality into (x−1)(x−3)⋅(x−2)(x−3)≤0  ⟺  (x−1)(x−2)(x−3)2≤0(x - 1)(x - 3) \cdot (x - 2)(x - 3) \le 0 \iff (x - 1)(x - 2)(x - 3)^2 \le 0.

Since (x−3)2≥0(x - 3)^2 \ge 0 for all real xx, the inequality holds when (x−1)(x−2)≤0(x - 1)(x - 2) \le 0 (giving [1,2][1, 2]) or when (x−3)2=0(x - 3)^2 = 0 (giving x=3x = 3).

The complete solution set is [1,2]∪{3}[1, 2] \cup \{3\}, containing the integers {1,2,3}\{1, 2, 3\} whose sum is 1+2+3=61 + 2 + 3 = 6.

Try real CAT questions

Study this concept with interactive practice in PrepSoc.

Open the concept library