Fact-checked by Grok 2 weeks ago

Power of three

In , a power of three is a number of the form $3^n, where n is an , obtained through with base 3 and exponent n. This operation represents repeated of 3 by itself n times when n is positive, yielding 1 when n=0, and reciprocals $1/3^{|n|} when n is negative. The sequence of powers of three, often considered for non-negative integers n, begins as 1, 3, 9, 27, 81, 243, 729, 2187, and continues indefinitely, with each subsequent term obtained by multiplying the previous by 3. These numbers exhibit properties such as being odd for all n \geq 1 and serving as perfect totient numbers, meaning the sum from k=1 to $3^n of \phi(k) equals $3^n. In base-10 representation, powers of three grow exponentially, with $3^{10} = 59049 and $3^{20} \approx 3.48 \times 10^9, illustrating their rapid increase. Powers of three play key roles in several areas of mathematics and . In , every integer has a unique representation in , a using digits -1, 0, and 1 as coefficients of powers of three, enabling efficient arithmetic without a separate . This system contrasts with standard (base 3 with digits 0-2) and has historical applications in early computing designs, such as ternary logic circuits. Additionally, powers of three appear in contexts, such as counting certain tree structures or in algorithms for sequence analysis, and in for partitioning problems involving multiples of 3.

Definition and Fundamentals

Definition

In mathematics, a power of three is defined as a number of the form $3^n, where n is an integer and the base 3 is raised to the exponent n through exponentiation. Exponentiation in this context represents the process of multiplying the base by itself n times when n is positive; for instance, $3^3 = 3 \times 3 \times 3 = 27. This sequence, known as the powers of 3, is cataloged in the Online Encyclopedia of Integer Sequences (OEIS) as A000244. The first ten terms of the sequence are:
$3^0 = 1,
$3^1 = [3](/page/3),
$3^2 = 9,
$3^3 = [27](/page/27),
$3^4 = 81,
$3^5 = 243,
$3^6 = [729](/page/729),
$3^7 = [2187](/page/21-87),
$3^8 = 6561,
$3^9 = 19683.
For completeness, powers of three extend to negative exponents, where $3^{-n} = \frac{1}{3^n} for positive n, resulting in fractions like $3^{-1} = \frac{1}{[3](/page/3)} and $3^{-2} = \frac{1}{9}. However, the concept primarily emphasizes non-negative exponents in contexts. Powers of three form the basis for place values in the .

Notation and Sequence

In , powers of three are denoted using exponential notation as $3^n, where n is an and the exponent appears as a superscript to the right of the 3. This compact form represents the product of n factors of 3 when n is positive, with $3^0 = 1 by convention. In computational and programming contexts, such as in C++ or , the operation is commonly expressed using the power function pow(3, n). The infinite sequence of powers of three, given by a_n = 3^n for n = 0, 1, 2, \dots, is cataloged as A000244 in the (OEIS). This sequence is strictly increasing for n \geq 0, as each term is precisely three times the preceding one, resulting in . Relative to the sequence of powers of two ($2^n), powers of three grow more rapidly owing to the larger base, yielding values that surpass those of $2^n for sufficiently large n. Higher powers of three can be computed through iterative multiplication, initializing with $3^0 = 1 and successively multiplying by 3 for each increment in the exponent: $3^k = 3 \times 3^{k-1}. For approximating large exponents without exact computation, the properties of logarithms provide a useful tool; specifically, \log_{10}(3^n) = n \log_{10} 3 \approx n \times 0.4771, which estimates the number of digits and magnitude as $3^n \approx 10^{n \times 0.4771}. The table below lists exact values of $3^n for n from 0 to 20, beyond which scientific notation is practical for representation.
n$3^n
01
13
29
327
481
5243
6729
72,187
86,561
919,683
1059,049
11177,147
12531,441
131,594,323
144,782,969
1514,348,907
1643,046,721
17129,140,163
18387,420,489
191,162,261,467
203,486,784,401

Mathematical Properties

Arithmetic and Algebraic Properties

The powers of three, expressed as $3^n for non-negative integers n, follow the standard laws of exponents in arithmetic operations involving multiplication and division. The product of two such powers is given by $3^m \times 3^n = 3^{m+n}, reflecting the additive property of exponents when bases are identical. Similarly, division yields $3^m / 3^n = 3^{m-n} provided m \geq n, as this simplifies the repeated multiplication inherent in the definition of exponents. Addition and of distinct powers of three do not admit simple closed-form expressions analogous to those for and division. However, the sum of consecutive powers starting from $3^0 = 1 up to $3^n constitutes a finite , with the formula S = \sum_{k=0}^n 3^k = \frac{3^{n+1} - 1}{2}. This summation leverages the common of 3 and derives from multiplying the series by the ratio and subtracting to isolate the result. Key algebraic identities further characterize powers of three. The difference $3^n - 1 factors as $3^n - 1 = (3 - 1) \sum_{k=0}^{n-1} 3^k = 2 \left( \frac{[3^n - 1](/page/N+1)}{2} \right), where the parenthetical is the up to n-1. Additionally, expressing $3^n as (1 + 2)^n allows expansion via the : $3^n = \sum_{k=0}^n \binom{n}{k} 2^k, which decomposes the power into terms weighted by binomial coefficients. The sequence $3^n demonstrates exponential growth, with base 3 exceeding e^1 \approx 2.718, corresponding to a natural logarithm \ln 3 > 1. This rapid expansion outpaces polynomial growth rates, such as linear O(n) or quadratic O(n^2), as the function scales multiplicatively with each increment in n.

Number-Theoretic Properties

Powers of three, expressed as $3^n for non-negative integers n, possess distinct number-theoretic characteristics rooted in their prime base. Excluding $3^0 = 1, all powers of three are odd integers, as the base 3 is odd and odd numbers raised to positive integer powers remain odd. In terms of primality, $3^1 = 3 is prime, while higher powers $3^n for n > 1 are composite, factoring as $3 \times 3^{n-1}. For example, $3^2 = 9 = 3 \times 3. Regarding divisibility, $3^n is divisible by $3^k whenever k \leq n, since $3^n = 3^k \cdot 3^{n-k}. Additionally, the relation to s arises in the factorization of $3^n - 1 = \prod_{d \mid n} \Phi_d(3), where \Phi_d(x) denotes the d-th . Powers of three also belong to special classes of integers. For n \geq 1, every $3^n is a perfect totient number, satisfying \sum_{k=1}^c \phi_k(3^n) = 3^n, where \phi_1(m) = \phi(m) is , \phi_{k+1}(m) = \phi(\phi_k(m)), and c is the smallest integer such that \phi_c(3^n) = 1. This holds because the iterated totients form the sequence $2 \cdot 3^{n-1}, 2 \cdot 3^{n-2}, \dots, 2, 1, and their sum is $2 \sum_{j=0}^{n-1} 3^j + 1 = 3^n. The powers of three further form the basis for a Stanley sequence, where the sequence consists of all possible sums of distinct powers of three, which is 3-free (containing no three-term arithmetic progression). No power of three is a perfect number, as the sum of divisors function gives \sigma(3^n) = \frac{3^{n+1} - 1}{2}, and setting this equal to $2 \cdot 3^n yields $3^{n+1} - 1 = 4 \cdot 3^n, or $3^n (3 - 4) = 1, implying -3^n = 1, which is impossible for non-negative integer n. Similarly, the only among powers of three is $3^1 = 3 = 2^2 - 1; for n > 1, no $3^n is a Mersenne prime, as the equation $2^p - 3^n = 1 with p, n > 1 has no solutions by Mihăilescu's theorem on consecutive powers.

Applications in Discrete Mathematics

In Numeral Systems

In the ternary numeral system, also known as base-3, numbers are represented using the digits 0, 1, and 2, with each positional place value corresponding to successive powers of three: $3^0 = 1 for the units place, $3^1 = 3 for the threes place, $3^2 = 9 for the nines place, and so forth. This system allows any non-negative integer to be expressed as a sum of these weighted digits, where the coefficient for each power of three is between 0 and 2. For instance, the ternary representation $10_3 denotes $1 \times 3^1 + 0 \times 3^0 = 3 in decimal notation. A notable variant is , which employs the digits -1, 0, and 1 (commonly symbolized as \overline{1}, 0, and 1 or -, 0, +) to achieve a unique and unambiguous representation for every , including negatives, without the need for a separate or leading zeros. Here, the place values remain powers of three, functioning as weights for the coefficients -1, 0, or 1, enabling compact encoding of both positive and negative values. For example, the balanced ternary number $1\overline{1}_3 equals $1 \times 3^1 + (-1) \times 3^0 = 3 - 1 = 2 in . This structure ensures that every can be uniquely converted to a sum of distinct powers of three multiplied by these coefficients, facilitating efficient arithmetic operations like and directly on the digits. The historical roots of numeral representations trace back to ancient Chinese counting rods, used from the (circa 5th century BCE) through the 17th century CE, where rods arranged on a board allowed . In modern computing, finds application in error detection mechanisms, such as ternary Berger codes, which enable self-checking circuits to detect faults in and transmission by leveraging the for improved reliability over equivalents.

In Combinatorics and Graph Theory

In , the expression $3^n enumerates the signed subsets of an n-element set, where each element may be included with a positive (+), included with a negative (-), or omitted (assigned ). This count arises from the three choices available for each of the n elements, corresponding to the cardinality of the set of all functions from the n- set to \{-[1](/page/1), [0](/page/0), [1](/page/1)\}. Similarly, Hanner polytopes, a of polytopes constructed recursively via Cartesian products and polar duals starting from line segments, possess exactly $3^d faces in d, achieving the minimum number of faces among unconditional polytopes of that dimension. The expression $3^n also appears in basic enumerative problems, such as counting the total number of strings of length n, where each position can be one of three symbols (e.g., , , or 2). This follows directly from the $3^n possible assignments, one for each position. In the context of paths, $3^n counts the unrestricted walks of length n on the plane (or higher dimensions) using a step set of three directions, such as east (1,0), north (0,1), and northeast (1,1), since each step offers three independent choices without constraints on the endpoint. In , $3^{n/3} provides the Moon–Moser bound on the maximum number of maximal cliques in an n- graph, achieved when n is divisible by 3 by the complete multipartite graph K_{3,3,\dots,3} with n/3 parts of size 3 each; in this Turán graph T(n, n/3), the maximal cliques are precisely the transversals selecting one from each part, yielding $3^{n/3} such cliques. For example, when n=12, this construction on 12 vertices produces exactly $3^4 = 81 maximal cliques. The , a procedure for enumerating all maximal cliques in an undirected graph, has a worst-case of O(3^{n/3}) in its version with pivoting, matching the Moon–Moser bound on the output size and thus optimal up to polynomial factors for graphs achieving the maximum number of maximal cliques.

Applications in Geometry and Large-Scale Structures

In Fractal Geometry

In fractal , powers of three frequently appear in the iterative construction of self-similar fractals, where scaling factors of \frac{1}{3} or removal of middle thirds dictate the at each stage. For instance, the is generated by starting with the unit interval [0,1] and iteratively removing the open middle third of each remaining subinterval; after n iterations, this leaves $2^n closed subintervals, each of length \frac{1}{3^n}. The of the is \frac{\log 2}{\log 3}, reflecting the scaling ratio of 3 and multiplicity of 2 per iteration. Similarly, the begins with an and replaces each side with four segments of length one-third the original in each iteration, adding protrusions scaled by \frac{1}{3}; this process increases the perimeter while bounding an area that converges to a finite value. The curve's is \frac{\log 4}{\log 3}, arising from the factor of 4 in segment multiplication against the linear scaling of \frac{1}{3}. The Sierpinski triangle is constructed by subdividing an into four smaller congruent triangles using midpoints and removing the central one, repeating on the remaining three; at iteration n, this yields $3^n smallest upward-pointing triangles, each with side length \left( \frac{1}{2} \right)^n. Its is \frac{\log 3}{\log 2}, determined by the triplication of substructures at half-scale. In three dimensions, the starts with a divided into 27 smaller cubes of side \frac{1}{3} and removes the central cube and the six face centers, leaving 20 subcubes; after n iterations, it consists of $20^n cubes, each of side length \frac{1}{3^n}. The sponge's is \frac{\log 20}{\log 3}, capturing the 20-fold replication at one-third scale. These iterative processes highlight the role of powers of three in defining fractal scaling, where inverse powers \frac{1}{3^n} govern size reduction and positive powers like $3^n or $20^n count the proliferating substructures.

In Hyperbolic and Large Numbers

In hyperbolic geometry, structures involving powers of three arise in tilings and packings that leverage three-fold symmetry to model the exponential expansion of space. The order-3 apeirogonal tiling, denoted by the Schläfli symbol \{\infty, 3\}, is a regular paracompact tiling of the hyperbolic plane in which three apeirogons (infinite-sided polygons) meet at each vertex. This configuration reflects the negative curvature of hyperbolic space, where the number of elements—such as vertices or edge segments—at geodesic distance n from a central vertex grows exponentially, with the growth rate governed by the tiling's adjacency structure rooted in the vertex degree of 3. Such tilings provide a framework for understanding large-scale geometric arrangements, where the iterative replication of triangular vertex figures leads to hierarchical expansions analogous to powers of three in discrete models. Horoball packings in 3-space further illustrate this connection, as they represent optimal density arrangements in decompositions derived from with triangular facets. For instance, packings associated with certain Coxeter groups achieve high densities through symmetric distributions involving three-way branching at vertices, resulting in layered structures where the number of horoballs scales exponentially with depth, effectively incorporating factors of $3^n in the of successive layers for certain group actions. These packings highlight how powers of three contribute to bounding volumes and covering efficiencies in unbounded domains. Powers of three also underpin some of the largest finite numbers in , particularly through iterated in . G, named after , serves as an upper bound for the smallest dimension n such that any 2-coloring of the edges of the on the $2^n vertices of an n-dimensional guarantees a monochromatic K_4 consisting of four coplanar vertices with all six connecting edges the same color. This problem, a specific instance of multidimensional , ensures the existence of ordered substructures in large combinatorial systems. The bound G vastly exceeds simple , illustrating the scale required for such guarantees in high dimensions. The construction of employs , introduced by to compactly denote hyperoperations beyond . The notation defines a \uparrow b = a^b, a \uparrow\uparrow b = a \uparrow (a \uparrow\uparrow (b-1)) with a \uparrow\uparrow 1 = a, and extends to more arrows for higher iterations: $3 \uparrow\uparrow 2 = 3^3 = 27, \quad 3 \uparrow\uparrow 3 = 3^{27} = 7{,}625{,}597{,}484{,}987, \quad 3 \uparrow\uparrow 4 = 3 \uparrow^{3^{27}} 3. Tetration $3 \uparrow\uparrow n already surpasses $3^n dramatically for n > 2, as it stacks exponents iteratively. Graham's number is then g_{64}, where g_1 = 3 \uparrow\uparrow\uparrow\uparrow 3 (three up-arrows between three 3's, equivalent to a power tower of 3's of immense height), and g_k = 3 \uparrow^{g_{k-1}} 3 for k \geq 2, with the superscript indicating the number of up-arrows. This 64-fold iteration yields a number incomprehensible in scale, where even the number of digits defies standard computation. Tetration and higher operations with base 3 thus provide the explosive growth needed to bound multicolored Ramsey numbers like R(3,3,3) in higher dimensions. This construction originated in the context of a 1971 paper by Graham and Bruce Rothschild, which proved a generalized for n- sets, establishing that sufficiently large sets partitioned into finitely many classes contain monochromatic combinatorial lines. Applying this to edge colorings yields the theoretical foundation for the bound, with the iterated powers of three ensuring the result holds across escalating dimensional complexities. The paper's insights into parameter words and partitions directly inform the problem, marking a seminal contribution to combinatorial bounds on large-scale structures.

References

  1. [1]
    Basic rules for exponentiation - Math Insight
    The power of power rule (4) allows us to define fractional exponents. For example, rule (4) tells us that 91/2=(32)1/2=32⋅1/2=31=3. Taking a number to the power ...Missing: three | Show results with:three
  2. [2]
    Exponentiation
    In Definition 1.31, we introduced the concept of multiplication as repeated addition, and we build upon that idea here. We define exponentiation as repeated ...
  3. [3]
    A000244 - OEIS
    Sep 25, 2025 · All powers of 3 are perfect totient numbers (A082897), since phi(3^n) = 2 * 3^(n - 1) for n > 0, and thus Sum_{i = 0..n} phi(3^i) = 3^n. - ...
  4. [4]
    Third Base | American Scientist
    The balanced ternary counting sequence begins: 0, 1, 11, 10, 11, 111, 110, 111. Going in the opposite direction, the first few negative numbers are 1, 11, 10, 1 ...
  5. [5]
    Exponentiation - Properties, Definition, Formula, Examples - Cuemath
    Exponentiation is the process of doing repeated multiplication by writing the given expression in the form of ab, where a is the base and b is the exponent.
  6. [6]
    Basic rules for exponentiation - Math Insight
    We can call this “x raised to the power of n,” “x to the power of n,” or simply “x to the n.” Here, x is the base and n is the exponent or the power.
  7. [7]
    Ternary numeral system - OeisWiki
    May 30, 2012 · The ternary numeral system (base 3) is a place-value notation for numbers using the powers of 3 rather than the powers of 10.Missing: three | Show results with:three
  8. [8]
    1.2: Exponents and Scientific Notation - Mathematics LibreTexts
    Dec 26, 2024 · When using the power rule, a term in exponential notation is raised to a power. In this case, you multiply the exponents. Table ...
  9. [9]
    Exponents - Math is Fun
    Example: 53 = 5 × 5 × 5 = 125. In words: 53 could be called "5 to the third power", "5 to the power 3" or simply "5 cubed" ; Example: 24 = 2 × 2 × 2 × 2 = 16. In ...
  10. [10]
    Write an iterative O(Log y) function for pow(x, y) - GeeksforGeeks
    Jul 23, 2025 · Step 1: Start the function with the base and exponent as input parameters. Step 2: Check if the exponent is equal to zero, return 1. Step 3: ...
  11. [11]
    Intro to exponential functions | Algebra (video) - Khan Academy
    Mar 19, 2014 · So let's just write an example exponential function here. So let's say we have y is equal to 3 to the x power. Notice, this isn't x to the third power, this is ...
  12. [12]
    15.3 Fast Exponentiation
    Now we look at example of computing powers in a group first using repeated multiplication and then using the method where we first write our exponent as a sum ...
  13. [13]
    Exponent properties review (article) | Khan Academy
    Review the common properties of exponents that allow us to rewrite powers in different ways. For example, x²⋅x³ can be written as x⁵.
  14. [14]
    Geometric Sequences and Sums - Math is Fun
    In a Geometric Sequence each term is found by multiplying the previous term by a constant. Example: 1, 2, 4, 8, 16, 32, 64, 128, 256, ...
  15. [15]
    Geometric Series -- from Wolfram MathWorld
    A geometric series sum_(k)a_k is a series for which the ratio of each two consecutive terms a_(k+1)/a_k is a constant function of the summation index k.<|separator|>
  16. [16]
    Intro to the Binomial Theorem (video) - Khan Academy
    Aug 4, 2019 · The Binomial theorem tells us how to expand expressions of the form (a+b)ⁿ, for example, (x+y)⁷. The larger the power is, the harder it is to expand ...
  17. [17]
    An Intuitive Guide To Exponential Functions & e - BetterExplained
    Because of the magic of exponents, we can avoid having two powers and just multiply rate and time together in a single exponent. The big secret: e merges rate ...
  18. [18]
    proof that all powers of 3 are perfect totient numbers - PlanetMath
    Mar 22, 2013 · proof that all powers of 3 are perfect totient numbers. Given an integer x>0 x > 0 , it is always the case that. 3x= ...
  19. [19]
    Ternary -- from Wolfram MathWorld
    Ternary ; 2^2, = 11_3 ; 2^3, = 22_3 ; 2^4, = 121_3 ; 2^5, = 1012_3 ; 2^6, = 2101_3.
  20. [20]
    Douglas W. Jones on Ternary Arithmetic - University of Iowa
    Unsigned and 3's-complement ternary arithmetic uses the values 0, 1, and 2 for each digit, while in balanced ternary uses –1, 0 and +1.
  21. [21]
    Chinese numerals - MacTutor History of Mathematics
    However a second form of Chinese numerals began to be used from the 4th century BC when counting boards came into use. A counting board consisted of a checker ...
  22. [22]
    Self-checking principle and design of ternary Berger code - Nature
    Apr 22, 2025 · Conclusion. This paper presents an error detection mode based on ternary logic Berger code. It can be used to detect errors in data storage and ...Missing: Setun | Show results with:Setun
  23. [23]
    [PDF] AC.pdf - Analytic Combinatorics
    Analytic combinatorics aims to enable precise quantitative predictions of the proper- ties of large combinatorial structures. The theory has emerged over ...
  24. [24]
    [PDF] arXiv:1906.01469v3 [math.CO] 26 Feb 2020
    Feb 26, 2020 · In this paper, we investigate unconditional lattice polytopes with respect to geometric, combinatorial, and algebraic properties. In particular, ...Missing: original | Show results with:original
  25. [25]
    [PDF] Chapter 10 - Lattice Path Enumeration - Universität Wien
    In particular, the importance of lattice path enumeration in non-parametric statistics seems to explain that the only books which are entirely devoted to ...
  26. [26]
    [PDF] On cliques in graphs - User Web Pages
    ABSTRACT. A clique is a maximal complete subgraph of a graph. The maximum number of cliques possible in a graph with n nodes is determined. Also, bounds are.
  27. [27]
    [PDF] The worst-case time complexity for generating all maximal cliques ...
    This is the best one could hope for as a function of n, since there exist up to 3n/3 maximal cliques in a graph with n vertices as shown by Moon and Moser [17].
  28. [28]
    Sierpiński Sieve -- from Wolfram MathWorld
    The Sierpiński sieve is a fractal described by Sierpiński in 1915 and appearing in Italian art from the 13th century (Wolfram 2002, p. 43).<|control11|><|separator|>
  29. [29]
    Menger Sponge -- from Wolfram MathWorld
    ... fractional volume after the n th iteration, then. N_n, = 20^n. (1). L_n, = (1/3)^n=3^(-n). (2). V_n, = L_n^3N_n=((20)/(27))^n. (3). The capacity dimension is ...Missing: cubes | Show results with:cubes
  30. [30]
    Structure and Visualization of Optimal Horoball Packings in $3 - arXiv
    Jan 14, 2016 · Four packings of hyperbolic 3-space are known to yield the optimal packing density of 0.85328\dots. They are realized in the regular tetrahedral ...Missing: powers | Show results with:powers
  31. [31]
    [PDF] The Optimal Ball and Horoball Packings to the Coxeter Honeycombs ...
    In [18] we investigated the regular Coxeter honeycombs and their optimal ball and horoball packings in the hyperbolic space H3. These Coxeter tilings are the.Missing: powers | Show results with:powers
  32. [32]
    Graham's Number -- from Wolfram MathWorld
    is the so-called Knuth up-arrow notation. g_(64) is often cited as the largest number that has ever been put to practical use (Exoo 2003). In chained arrow ...<|control11|><|separator|>
  33. [33]
    Knuth Up-Arrow Notation -- from Wolfram MathWorld
    Knuth's up-arrow notation is a notation invented by Knuth (1976) to represent large numbers in which evaluation proceeds from the right.Missing: source | Show results with:source