{
  "id": "infinite-limit",
  "title": "Infinite Limit",
  "lang": "en",
  "source_lang": "en",
  "summary": "An infinite limit describes a function or sequence whose values eventually exceed every positive bound or fall below every negative bound.",
  "category": "mathematics",
  "subcategory": "analysis",
  "type": "concept",
  "aliases": [
    "Limit equal to infinity",
    "Divergence to infinity"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/infinite-limit/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/infinite-limit/",
      "json": "https://aiwiki.page/zh/infinite-limit.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "An infinite limit describes the behavior of a function or sequence whose values tend to \\(+\\infty\\) or \\(-\\infty\\), rather than approaching a finite real number. In calculus and mathematical analysis, the statement \\(\\lim_{x\\to a}f(x)=+\\infty\\) means that \\(f(x)\\) exceeds every prescribed positive threshold whenever \\(x\\) is sufficiently close to \\(a\\), excluding \\(a\\) itself. The corresponding negative infinite limit means that the values eventually lie below every prescribed negative threshold. These statements describe precise forms of divergence, not convergence to an ordinary real number."
    },
    {
      "heading": "Definitions",
      "text": "Let \\(f:D\\to\\mathbb R\\), where \\(D\\subseteq\\mathbb R\\), and suppose that \\(a\\) is an accumulation point of \\(D\\): points of \\(D\\) distinct from \\(a\\) occur arbitrarily close to \\(a\\).\n\nThe notation\n\n\\[\\lim_{x\\to a}f(x)=+\\infty\\]\n\nmeans that for every \\(M>0\\), there exists \\(\\delta>0\\) such that\n\n\\[x\\in D,\\quad 0<|x-a|<\\delta \\quad\\Longrightarrow\\quad f(x)>M.\\]\n\nSimilarly,\n\n\\[\\lim_{x\\to a}f(x)=-\\infty\\]\n\nmeans that for every \\(M>0\\), there exists \\(\\delta>0\\) such that\n\n\\[x\\in D,\\quad 0<|x-a|<\\delta \\quad\\Longrightarrow\\quad f(x)<-M.\\]\n\nThe restriction \\(x\\in D\\) makes these definitions relative to the domain of the function. The exclusion \\(x\\ne a\\) means that \\(f(a)\\), if defined, has no effect on the limit.\n\nThe crucial quantifier is “for every”: passing one large threshold is insufficient. Once any threshold has been chosen, all sufficiently nearby domain points must produce values beyond it.\n\nExample and proof\n\nFor\n\n\\[f(x)=\\frac{1}{(x-a)^2},\\]\n\nthe limit is\n\n\\[\\lim_{x\\to a}\\frac{1}{(x-a)^2}=+\\infty.\\]\n\nGiven \\(M>0\\), choose \\(\\delta=1/\\sqrt M\\). Then\n\n\\[0<|x-a|<\\delta \\quad\\Longrightarrow\\quad \\frac{1}{(x-a)^2}>M.\\]\n\nThis directly verifies the definition."
    },
    {
      "heading": "One-sided infinite limits",
      "text": "A right-hand infinite limit restricts the approach to \\(x>a\\). For example,\n\n\\[\\lim_{x\\to a^+}f(x)=+\\infty\\]\n\nmeans that for every \\(M>0\\), some \\(\\delta>0\\) satisfies\n\n\\[x\\in D,\\quad a<x<a+\\delta \\quad\\Longrightarrow\\quad f(x)>M.\\]\n\nLeft-hand limits use \\(a-\\delta<x<a\\). Negative infinite limits are defined by replacing \\(f(x)>M\\) with \\(f(x)<-M\\).\n\nWhen the domain has points approaching \\(a\\) from both sides, a two-sided limit equals \\(+\\infty\\) exactly when both one-sided limits equal \\(+\\infty\\). The analogous statement holds for \\(-\\infty\\).\n\nFor instance,\n\n\\[\\lim_{x\\to0^+}\\frac1x=+\\infty, \\qquad \\lim_{x\\to0^-}\\frac1x=-\\infty.\\]\n\nConsequently, \\(1/x\\) has no two-sided limit at zero, even if signed infinite limits are admitted. By contrast,\n\n\\[\\lim_{x\\to0}\\frac1{x^2}=+\\infty.\\]\n\nA line \\(x=a\\) is commonly called a vertical asymptote of the graph when at least one of the one-sided limits at \\(a\\) is \\(+\\infty\\) or \\(-\\infty\\). This is a statement about nearby behavior; changing or assigning the value at \\(a\\) does not remove the asymptote."
    },
    {
      "heading": "Infinite limits versus limits at infinity",
      "text": "An infinite limit concerns the output values. A limit at infinity concerns the input variable. The two notions can occur separately or together.\n\nExamples include\n\n\\[\\lim_{x\\to+\\infty}\\frac1x=0, \\qquad \\lim_{x\\to+\\infty}x^2=+\\infty.\\]\n\nThe first is a finite limit at infinity; the second is an infinite limit at infinity.\n\nFormally,\n\n\\[\\lim_{x\\to+\\infty}f(x)=+\\infty\\]\n\nmeans that for every \\(M>0\\), there exists \\(A\\in\\mathbb R\\) such that\n\n\\[x\\in D,\\quad x>A \\quad\\Longrightarrow\\quad f(x)>M.\\]\n\nHere the domain is assumed to be unbounded above. Limits as \\(x\\to-\\infty\\) instead impose \\(x<A\\).\n\nFor a polynomial, the leading term determines its infinite-end behavior. For example,\n\n\\[\\lim_{x\\to+\\infty}(3x^3-5x+1)=+\\infty, \\qquad \\lim_{x\\to-\\infty}(3x^3-5x+1)=-\\infty.\\]"
    },
    {
      "heading": "Sequences and the distinction from unboundedness",
      "text": "A real sequence \\((a_n)\\) tends to \\(+\\infty\\) if, for every \\(M>0\\), there exists an index \\(N\\) such that\n\n\\[n\\ge N\\quad\\Longrightarrow\\quad a_n>M.\\]\n\nIt tends to \\(-\\infty\\) if the eventual inequality is \\(a_n<-M\\).\n\nThus \\(a_n=n^2\\) tends to \\(+\\infty\\), while \\(a_n=(-1)^n n\\) tends to neither signed infinity. The latter sequence has arbitrarily large positive and negative terms.\n\nAn infinite limit is therefore stronger than unboundedness. A sequence is unbounded above if it has terms above every threshold; tending to \\(+\\infty\\) requires every sufficiently late term to exceed each threshold. Likewise, a function may be unbounded in every neighborhood of a point without having an infinite limit there. For example,\n\n\\[f(x)=\\frac{\\sin(1/x)}{x^2}\\]\n\ntakes arbitrarily large positive and negative values near zero.\n\nInfinite limits do not require monotonicity. The sequence\n\n\\[a_n=n+2(-1)^n\\]\n\ntends to \\(+\\infty\\) despite repeatedly decreasing between consecutive terms. Conversely, a monotone increasing sequence that is unbounded above necessarily tends to \\(+\\infty\\).\n\nThe sequence formulation also characterizes function limits: \\(\\lim_{x\\to a}f(x)=+\\infty\\) if and only if \\(f(x_n)\\to+\\infty\\) for every sequence of domain points with \\(x_n\\ne a\\) and \\(x_n\\to a\\)."
    },
    {
      "heading": "Calculation rules and indeterminate forms",
      "text": "Many familiar limit rules extend to infinite limits, provided that signs and exceptional cases are handled correctly. At the same limiting point:\n\n- If \\(f\\to+\\infty\\) and \\(g\\to L\\in\\mathbb R\\), then \\(f+g\\to+\\infty\\).\n- If \\(f\\to+\\infty\\) and \\(g\\to c>0\\), then \\(fg\\to+\\infty\\).\n- If \\(f\\to+\\infty\\) and \\(g\\to c<0\\), then \\(fg\\to-\\infty\\).\n- If \\(f\\to+\\infty\\) or \\(f\\to-\\infty\\), then \\(1/f\\to0\\).\n- If \\(g\\to0\\) through positive values, then \\(1/g\\to+\\infty\\); approaching zero through negative values gives \\(-\\infty\\).\n\nA useful comparison principle states that if \\(f(x)\\ge g(x)\\) eventually and \\(g(x)\\to+\\infty\\), then \\(f(x)\\to+\\infty\\).\n\nExpressions conventionally written as\n\n\\[\\infty-\\infty,\\qquad 0\\cdot\\infty,\\qquad \\frac{\\infty}{\\infty}\\]\n\nare indeterminate forms. They do not specify a limit without further information. For example, as \\(x\\to+\\infty\\),\n\n\\[x^2-x\\to+\\infty,\\qquad x-x=0,\\qquad x-x^2\\to-\\infty,\\]\n\nalthough all three differences have the symbolic form \\(\\infty-\\infty\\).\n\nSuch expressions require analysis of the original functions, using methods such as algebraic simplification, comparison, or asymptotic expansions. Infinity is not a real number that can be substituted into ordinary arithmetic."
    },
    {
      "heading": "Extended-real interpretation",
      "text": "In ordinary real analysis, a sequence tending to infinity is described as divergent because it does not converge to a real number. On the extended real number line,\n\n\\[\\overline{\\mathbb R}=\\mathbb R\\cup\\{-\\infty,+\\infty\\},\\]\n\nthe same behavior can be treated as convergence to an added endpoint.\n\nThe topology of this space makes sets of the form \\((M,+\\infty]\\) neighborhoods of \\(+\\infty\\). Convergence to \\(+\\infty\\) therefore means eventually entering every such neighborhood, exactly as in the threshold definition.\n\nThis interpretation does not turn the extended real line into a field: expressions such as \\(+\\infty-(+\\infty)\\) remain undefined. Also, \\(|f(x)|\\to+\\infty\\) need not imply either signed infinite limit, because the sign may keep changing."
    },
    {
      "heading": "Role in improper integrals and series",
      "text": "An improper integral is defined through a limit of ordinary integrals. For example,\n\n\\[\\int_1^\\infty \\frac{1}{x}\\,dx = \\lim_{R\\to+\\infty}\\int_1^R\\frac{1}{x}\\,dx = \\lim_{R\\to+\\infty}\\log R = +\\infty.\\]\n\nThe integral is therefore said to diverge to \\(+\\infty\\).\n\nAn unbounded integrand does not by itself imply a divergent integral. For example,\n\n\\[\\lim_{x\\to0^+}\\frac1{\\sqrt{x}}=+\\infty, \\qquad \\int_0^1\\frac1{\\sqrt{x}}\\,dx=2.\\]\n\nThe rate of growth near the singular point matters.\n\nSimilarly, an infinite series diverges to \\(+\\infty\\) when its partial sums tend to \\(+\\infty\\). The terms themselves need not grow: the harmonic series has terms \\(1/n\\) tending to zero, while its partial sums tend to \\(+\\infty\\)."
    }
  ],
  "links": [
    "mathematical-function",
    "real-number",
    "calculus",
    "mathematical-analysis",
    "domain-of-a-function",
    "limit",
    "polynomial",
    "monotonic-function",
    "asymptotic-expansion",
    "topology",
    "field-mathematics",
    "improper-integral",
    "infinite-series"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}