Unit-generated orders of real quadratic fields I. Class number bounds
Pith reviewed 2026-05-16 23:14 UTC · model grok-4.3
The pith
Unit-generated orders in real quadratic fields have class numbers satisfying log |Cl(O)| ∼ log(½ |Δ(O)|) as |Δ(O)| grows large.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
Unit-generated orders are those of the form O = Z[ε] where ε is a unit of the quadratic field. For real quadratic fields these orders are parametrized by O_n^+ (Δ = n²-4, n≥3) and O_n^- (Δ = n²+4, n≥1). The central theorem states that the (wide or narrow) class number satisfies log |Cl(O)| ∼ log(½ |Δ(O)|) as |Δ(O)| → ∞, obtained by applying Hua's class-number result to this explicit family. Direct corollaries are that only finitely many unit-generated orders have class number one (and all are classified) and that only finitely many have 2-torsion class groups (with lists given up to Δ ≤ 10^10).
What carries the argument
The two-parameter families of unit-generated orders O_n^± whose discriminants are exactly n²-4 and n²+4, to which Hua's class-number estimate is applied directly.
If this is right
- Only finitely many unit-generated quadratic orders have class number one.
- All unit-generated orders of class number one are explicitly classified.
- Only finitely many unit-generated orders have 2-torsion class groups (wide or narrow).
- Explicit lists of all 2-torsion examples exist for |Δ| ≤ 10^10 and are conjecturally complete.
Where Pith is reading between the lines
- These orders form an explicit infinite family in which the generic logarithmic growth of class numbers can be verified directly.
- The finiteness statements may extend to other parametrized families of orders with bounded generator rank.
- Computational checks for discriminants beyond 10^10 would test whether the provided 2-torsion lists remain exhaustive.
- The work supplies concrete data that can be used to compare class-number statistics between unit-generated orders and maximal orders.
Load-bearing premise
Hua's class-number result applies directly to the specific discriminants n² ± 4 without additional restrictions or larger error terms.
What would settle it
Discovery of a sequence of unit-generated orders O_n with |Δ(O_n)| → ∞ for which log |Cl(O_n)| grows strictly slower than any positive multiple of log |Δ(O_n)|, for instance with |Cl(O_n)| bounded.
read the original abstract
Unit-generated orders of a quadratic field are orders of the form $\mathcal{O} = \mathbb{Z}[\varepsilon]$, where $\varepsilon$ is a unit in the quadratic field. If the order $\mathcal{O}$ is a maximal order of a real quadratic field, then the quadratic number field is necessarily of a restricted form, being of narrow Richaud--Degert type. However, every real quadratic field contains infinitely many distinct unit-generated orders. They are parametrized as $\mathcal{O} = \mathcal{O}_{n}^{\pm}$ having quadratic discriminants $\Delta(\mathcal{O}) = \Delta_{n}^{+} = n^2 - 4$ (for $n \geq 3$) and $\Delta(\mathcal{O}) = \Delta_{n}^{-} = n^2 + 4$ (for $n \geq 1$). We show the (wide or narrow) class numbers of unit-generated orders satisfy $\log \left|{\rm Cl}(\mathcal{O})\right| \sim \log \frac{1}{2}\left|\Delta(\mathcal{O})\right|$ as $\left|\Delta(\mathcal{O})\right| \to \infty$, using a result of L.-K. Hua. We deduce that there are finitely many unit-generated quadratic orders of class number one and finitely many unit-generated quadratic orders whose class group is $2$-torsion. We classify all unit-generated real quadratic orders having class number one. We provide numerical lists of quadratic unit-generated orders whose class groups are $2$-torsion for $\Delta \leq 10^{10}$, for both wide and narrow class groups. These lists are conjecturally complete for all $\Delta$.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper parametrizes unit-generated orders O_n^± in real quadratic fields by discriminants Δ_n^+ = n²-4 (n≥3) and Δ_n^- = n²+4 (n≥1), claims that their (wide or narrow) class numbers satisfy log |Cl(O)| ∼ log(|Δ(O)|/2) as |Δ|→∞ via a result of L.-K. Hua, deduces finiteness of class-number-one and 2-torsion examples, classifies all class-number-one cases, and supplies explicit lists of 2-torsion examples for |Δ|≤10^10.
Significance. A correct asymptotic for class numbers in this explicit family would immediately yield the stated finiteness theorems and furnish a complete classification up to any fixed bound. The numerical lists up to 10^10 are obtained by direct computation on the parametrization and would remain useful even if the asymptotic statement is revised.
major comments (1)
- [Abstract] Abstract and introduction: the claimed asymptotic log |Cl(O)| ∼ log(|Δ(O)|/2) cannot hold. For O = ℤ[ε] with ε the fundamental unit of norm ±1, one has |ε| ∼ n and thus R(O) = log|ε| ∼ (1/2) log|Δ|. The Dirichlet class-number formula then gives h(O) R(O) ∼ sqrt(|Δ|) L(1,χ_Δ) with L(1,χ_Δ) ≪ log|Δ| on average, implying log h(O) ∼ (1/2) log|Δ| − log log|Δ|, not log|Δ|. The application of Hua’s result therefore does not produce the stated growth rate for this special family.
Simulated Author's Rebuttal
We thank the referee for their careful reading of the manuscript and for identifying the issue with the asymptotic statement. We respond to the major comment below.
read point-by-point responses
-
Referee: [Abstract] Abstract and introduction: the claimed asymptotic log |Cl(O)| ∼ log(|Δ(O)|/2) cannot hold. For O = ℤ[ε] with ε the fundamental unit of norm ±1, one has |ε| ∼ n and thus R(O) = log|ε| ∼ (1/2) log|Δ|. The Dirichlet class-number formula then gives h(O) R(O) ∼ sqrt(|Δ|) L(1,χ_Δ) with L(1,χ_Δ) ≪ log|Δ| on average, implying log h(O) ∼ (1/2) log|Δ| − log log|Δ|, not log|Δ|. The application of Hua’s result therefore does not produce the stated growth rate for this special family.
Authors: We agree with the referee that the asymptotic as written cannot hold. The regulator of these orders satisfies R(O) ∼ (1/2) log |Δ(O)|, and the class-number formula for orders then implies that the growth of |Cl(O)| is determined by sqrt(|Δ|) L(1, χ_Δ) / R(O). We had intended to invoke Hua's result to obtain a lower bound of the form |Cl(O)| ≫ log |Δ(O)| (or the precise logarithmic growth furnished by that theorem for this family), which remains sufficient for the finiteness statements on class-number-one orders and orders with 2-torsion class groups. The outer logarithm on the left-hand side of the claimed relation was included in error. We will revise the abstract, introduction, and the statement of the main asymptotic result to remove the extraneous logarithm and to clarify the precise application of Hua's theorem. The corrected statement will still tend to infinity with |Δ|, so the finiteness deductions, the classification of class-number-one cases, and the explicit lists up to |Δ| ≤ 10^10 remain valid and will be retained. revision: yes
Circularity Check
No significant circularity; derivation applies external Hua theorem to explicit parametrization
full rationale
The paper obtains the asymptotic log |Cl(O)| ∼ log(1/2 |Δ(O)|) by direct application of Hua's external class-number result to the explicitly parametrized family O_n^± with Δ_n^± = n² ± 4. Finite classifications and lists for class number one and 2-torsion cases are produced by direct computation on this parametrization, not by fitting parameters or renaming inputs as outputs. No load-bearing step reduces to self-definition, self-citation chains, or ansatz smuggling; the chain is independent of the claimed conclusion and rests on the external benchmark.
Axiom & Free-Parameter Ledger
axioms (1)
- domain assumption Hua's theorem on the growth of class numbers for quadratic orders applies to the family O_n^±
Reference graph
Works this paper leans on
-
[1]
D. M. Appleby, H. Yadsan-Appleby, and G. Zauner. Galois automorphisms of a symmetric measurement. Quantum Inf. Comput., 13(7-8):672–720, 2013
work page 2013
-
[2]
M. Appleby, I. Bengtsson, M. Grassl, M. Harrison, and G. McConnell. SIC-POVMs from Stark units: Prime dimensionsn 2 + 3.J. Math. Phys., 63(11):Paper No. 112205, 31 pp., 2022
work page 2022
-
[3]
M. Appleby, S. Flammia, G. McConnell, and J. Yard. SICs and algebraic number theory.Found. Phys., 47(8):1042–1059, 2017
work page 2017
-
[4]
M. Appleby, S. Flammia, G. McConnell, and J. Yard. Generating ray class fields of real quadratic fields via complex equiangular lines.Acta Arith., 192(3):211–233, 2020
work page 2020
-
[5]
M. Appleby, S. T. Flammia, and G. S. Kopp. A constructive approach to Zauner’s conjecture via the Stark conjectures, 2025. Preprint arXiv:2501.03970
-
[6]
E. Bach and J. Shallit.Algorithmic Number Theory, Volume 1: Efficient Algorithms. Foundations of Computing Series. MIT Press, Cambridge, MA, 1996
work page 1996
-
[7]
I. Bengtsson, M. Grassl, and G. McConnell. SIC-POVMs from Stark units: Dimensionsn 2 + 3 = 4p,p prime.J. Math. Phys., 66(8):Paper No. 082202, 38 pp., 2025
work page 2025
-
[8]
A. Bir´ o. Chowla’s conjecture.Acta Arith., 107(2):179–194, 2003
work page 2003
-
[9]
A. Bir´ o. Yokoi’s conjecture.Acta Arith., 106(1):85–104, 2003
work page 2003
- [10]
-
[11]
E. Caeiro and H. Darmon. The Heegner–Stark theorem and Stark–Heegner points.Essent. Number Theory, 4(1):67–99, 2025
work page 2025
-
[12]
S. Chowla and W. E. Briggs. On discriminants of binary quadratic forms with a single class in each genus.Canad. J. Math., 6:463–470, 1954
work page 1954
-
[13]
S. Chowla and J. Friedlander. Class numbers and quadratic residues.Glasgow Math. J., 17(1):47–52, 1976
work page 1976
-
[14]
Dedekind.Ueber die Anzahl der Ideal-Classen in den Verschiedenen Ordnungen Eines Endlichen K¨ orpers
R. Dedekind.Ueber die Anzahl der Ideal-Classen in den Verschiedenen Ordnungen Eines Endlichen K¨ orpers. Vieweg, 1877
- [15]
-
[16]
Dedekind.Theory of Algebraic Integers
R. Dedekind.Theory of Algebraic Integers. Cambridge Mathematical Library. Cambridge University Press, Cambridge, 1996. Translated from the 1877 French original by John Stillwell
work page 1996
-
[17]
G. Degert. ¨Uber die Bestimmung der Grundeinheit gewisser reell-quadratischer Zahlk¨ orper.Abh. Math. Sem. Univ. Hamburg, 22:92–97, 1958
work page 1958
-
[18]
P. G. L. Dirichlet. Une propri´ et´ e des formes quadratiques a d´ eterminant positif.J. de Math. Pures. Appl., pages 76–79, 1856
-
[19]
L. Euler. Extrait d’une lettre de. M. Euler ` a M. B´ eguelin (en Mai 1778).Nouveaux M´ emoirs de l’Acad´ emie des Sciences de Berlin 1776, pages 337–339, 1779
-
[20]
L. Euler. De formulis specieimxx+nyyad numeros primos explorados idoneis earumque mirabilibus proprietatibus.Nova acta acad. scien. Petropolitonae, 12(1794):22–46, 1801
-
[21]
G. Frei. Les nombres convenables de Leonhard Euler. InNumber theory (Besan¸ con), 1983–1984, Publ. Math. Fac. Sci. Besan¸ con, pages Exp. No. 1, 58 pp. Univ. Franche-Comt´ e, Besan¸ con, 1984
work page 1983
-
[22]
C. F. Gauss.Disquisitiones arithmeticae. Springer-Verlag, New York, 1986. Translated from the 1801 German original by Arthur A. Clarke
work page 1986
-
[23]
M. Grassl and A. J. Scott. Fibonacci-Lucas SIC-POVMs.J. Math. Phys., 58(12):122201, 13 pp., 2017. UNIT-GENERATED ORDERS I 21
work page 2017
-
[24]
F. Grube. Ueber einige Euler’sche S¨ atze aus der Theorie der quadratische Formen.Zeitschrift f¨ ur Math- ematik und Physik, 19:492–519, 1874
-
[25]
Halter-Koch.Quadratic Irrationals: An Introduction to Classical Number Theory
F. Halter-Koch.Quadratic Irrationals: An Introduction to Classical Number Theory. Pure and Applied Mathematics. CRC Press, Boca Raton, FL, 2013
work page 2013
-
[26]
H. Hasse. ¨Uber mehrklassige, aber eingeschlechtige reell-quadratische Zahlk¨ orper.Elem. Math., 20:49– 59, 1965
work page 1965
-
[27]
Hua.Introduction to Number Theory
L.-K. Hua.Introduction to Number Theory. Springer-Verlag, Berlin-New York, 1982. Translated from the 1957 Chinese original by Peter Shiu
work page 1982
-
[28]
E. Kani. Idoneal numbers and some generalizations.Ann. Sci. Math. Qu´ ebec, 35(2):197–227, 2011
work page 2011
-
[29]
F. Kawamoto and Y. Kishi. Continued fractions and the class number of real quadratic orders.Tokyo J. Math., 47(1):245–280, 2024
work page 2024
-
[30]
F. Kawamoto, Y. Kishi, and K. Tomita. On some subgroups of the ideal class group of real quadratic orders.Kyushu J. Math., 78(2):433–485, 2024
work page 2024
-
[31]
F. Kawamoto and K. Tomita. Continued fractions and Gauss’ class number problem for real quadratic fields.Tokyo J. Math., 35(1):213–239, 2012
work page 2012
-
[32]
G. S. Kopp. SIC-POVMs and the Stark conjectures.Int. Math. Res. Not. IMRN, 2021(18):13812–13838, 2021
work page 2021
-
[33]
G. S. Kopp and J. C. Lagarias. Unit-generated orders of real quadratic fields II. Class monoids and special ray class fields. Paper in preparation
-
[34]
G. S. Kopp and J. C. Lagarias. SIC-POVMs and orders of real quadratic fields, 2024. Preprint 2407.08048
work page internal anchor Pith review Pith/arXiv arXiv 2024
-
[35]
G. S. Kopp and J. C. Lagarias. Ray class groups and ray class fields for orders of number fields.Essent. Number Theory, 4(1):1–65, 2025
work page 2025
-
[36]
F. Lemmermeyer. The development of the principal genus theorem. InThe Shaping of Arithmetic after C. F. Gauss’s Disquisitiones Arithmeticae, pages 529–561. Springer, Berlin, 2007
work page 2007
-
[37]
S. Louboutin, R. A. Mollin, and H. C. Williams. Class groups of exponent two in real quadratic fields. InAdvances in number theory (Kingston, ON, 1991), Oxford Sci. Publ., pages 499–513. Oxford Univ. Press, New York, 1993
work page 1991
-
[38]
R. A. Mollin. Class number one criteria for real quadratic fields. I.Proc. Japan Acad. Ser. A Math. Sci., 63(4):121–125, 1987
work page 1987
-
[39]
R. A. Mollin. On the insolubility of a class of Diophantine equations and the nontriviality of the class numbers of related real quadratic fields of Richaud-Degert type.Nagoya Math. J., 105:39–47, 1987
work page 1987
-
[40]
R. A. Mollin.Quadratics. CRC Press Series on Discrete Mathematics and its Applications. CRC Press, Boca Raton, FL, 1996
work page 1996
-
[41]
R. A. Mollin. Class number one and prime-producing quadratic polynomials revisited.Canad. Math. Bull., 41(3):328–334, 1998
work page 1998
-
[42]
R. A. Mollin and H. C. Williams. Solution of the class number one problem for real quadratic fields of extended Richaud-Degert type (with one possible exception). InNumber theory (Banff, AB, 1988), pages 417–425. de Gruyter, Berlin, 1990
work page 1988
-
[43]
H. L. Montgomery and R. C. Vaughan.Multiplicative number theory. I. Classical theory, volume 97 of Cambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2007
work page 2007
-
[44]
Neukirch.Algebraic number theory, volume 322 ofGrundlehren der mathematischen Wissenschaften
J. Neukirch.Algebraic number theory, volume 322 ofGrundlehren der mathematischen Wissenschaften. Springer-Verlag, Berlin, 1999. Translated from the 1992 German original by Norbert Schappacher
work page 1999
-
[45]
C. Richaud. Sur la r´ esolution des ´ equationsx2 −Ay 2 =±1.Atti Accad. Pontif. Nuovi Lincei, pages 177–182, 1866
-
[46]
J. B. Rosser and L. Schoenfeld. Approximate formulas for some functions of prime numbers.Illinois J. Math., 6:64–94, 1962
work page 1962
-
[47]
C. L. Siegel. ¨Uber die classenzahl quadratischer zahlk¨ orper.Acta Arith., 1:83–86, 1935
work page 1935
-
[48]
H. M. Stark. The Gauss class-number problems. InAnalytic number theory, volume 7 ofClay Math. Proc., pages 247–256. Amer. Math. Soc., Providence, RI, 2007
work page 2007
-
[49]
H. Yokoi. Class-number one problem for certain kind of real quadratic fields. InProceedings of the international conference on class numbers and fundamental units of algebraic number fields (Katata, 1986), pages 125–137. Nagoya Univ., Nagoya, 1986. 22 GENE S. KOPP AND JEFFREY C. LAGARIAS
work page 1986
-
[50]
D. B. Zagier.Zetafunktionen und quadratische K¨ orper: Eine Einf¨ uhrung in die h¨ ohere Zahlentheorie. Hochschultext. Springer-Verlag, Berlin-New York, 1981. Department of Mathematics, Louisiana State University, Baton Rouge, LA, USA Email address:kopp@math.lsu.edu Department of Mathematics, University of Michigan, Ann Arbor, MI, USA Email address:lagari...
work page 1981
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.