Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

Embed Size (px)

Citation preview

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    1/178

    RESEARCH ON

    SMARANDACHE PROBLEMS

    IN NUMBER THEORY

    (Collected papers)

    Edited by

    ZHANG WENPENG

    Department of Mathematics

    Northwest University

    Xian, P. R. China

    Hexis

    2004

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    2/178

    RESEARCH ON

    SMARANDACHE PROBLEMS

    IN NUMBER THEORY(Collected papers)

    Edited by

    ZHANG WENPENG

    Department of Mathematics

    Northwest University

    Xian, P. R. China

    Hexis

    2004

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    3/178

    This book can be ordered in a paper bound reprint from:

    Books on Demand

    ProQuest Information & Learning

    (University of Microfilm International)300 N. Zeeb RoadP.O. Box 1346, Ann Arbor

    MI 48106-1346, USATel.: 1-800-521-0600 (Customer Service)

    http://wwwlib.umi.com/bod/search/basic

    Peer Reviewers:A. W. Vyawahare,H. O. D. Mathematics Department, M. M. College Of Science, Umred

    Road, Sakkardara, Nagpur University, Nagpur, PIN :- 440009, India.K. M. Purohit,H. O. D. Mathematics Department, V.M.V. Com., J.M.T. Arts & J.J.P.

    Science College, Wardhaman Nagar, Nagpur University, Nagpur, PIN : 440008 , India.Dr. (Mrs.) W.B.Vasantha Kandasamy, Department of Mathematics, Indian Institute of

    Technology, IIT Madras, Chennai - 600 036, India.

    Copyright 2004 by Hexis (Phoenix, USA) and Zhang Wenpeng, and Authors.

    Many books one can download from the Digital Library:http://www.gallup.unm.edu/~smarandache/eBooks-otherformats.htm

    ISBN: 1-931233-88-8

    Standard Address Number: 297-5092

    Printed in the United States of America

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    4/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    5/178

    Contents

    Dedication v

    Preface xi

    An arithmetic function and the primitive number of power

    1Zhang Wenpeng

    On the primitive numbers of power

    and

    -power roots 5Yi Yuan, Liang Fangchi

    Mean value on the pseudo-Smarandache squarefree function 9Liu Huaning, Gao Jing

    On the additive

    -th power complements 13Xu Zhefeng

    On the Smarandache pseudo-multiples of sequence 17Wang Xiaoying

    An arithmetic function and the divisor product sequences 21Zhang Tianping

    The Smarandache irrational root sieve sequences 27Zhang Xiaobeng, Lou Yuanbing

    A number theoretic function and its mean value 33Lv Chuan

    On the primitive numbers of power

    and its triangle inequality 37Ding Liping

    The additive analogue of Smarandache simple function 39Zhu Minhui

    On the

    -power complement sequence 43Yao Weili

    On the inferior and superior factorial part sequences 47Li Jie

    A number theoretic function and its mean value 49Gao Nan

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    6/178

    viii RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    On the generalized constructive set 53Gou su

    On the inferior and superior prime part 57Lou Yuanbing

    Identities on the

    -power complements 61Zhang Wenpeng

    On the asymptotic property of divisor function for additive complements 65Yi Yuan, Liang Fangchi

    Mean value on two Smarandache-type multiplicative functions 69

    Liu Huaning, Gao Jing

    On the Smarandache ceil function and the number of prime factors 73Xu Zhefeng

    On the mean value of an arithmetical function 77Wang Xiaoying

    Two asymptotic formulae on the divisor product sequences 81Zhang Tianping

    On the Smarandache pseudo-even number Sequence 85Zhang Xiaobeng, Lou Yuanbing

    On the mean value of an arithmetical function 89Lv Chuan

    An arithmetical function and its cubic complements 93Ding Liping

    On the symmetric sequence and its some properties 97Zhu Minhui

    The additive analogue of Smarandache function 99Yao Weili

    An asymptotic formula on Smarandache ceil function 103Li Jie

    A hybrid number theoretic function and its mean value 107Gao Nan

    On the Smarandache pseudo-number 111Lou Yuanbing

    Several asymptotic formulae on a new arithmetical function 115Guo Jinbao and He Yanfeng

    On the Smarandache function and the

    -th roots of a positive integer 119 Li Hailong and Zhao Xiaopeng

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    7/178

    Contents ix

    On a dual of the Pseudo Smarandache Function and its Asymptotic Formula 123Liu Duansen and Yang Cundian

    The primitive numbers of power

    and its asymptotic property 129Liang Fangchi, Yi Yuan

    Some Asymptotic properties involving the Smarandache ceil function 133 He Xiaolin and Guo Jinbao

    Asymptotic formulae of Smarandache-type multiplicative Functions 139Yang Cundian and Li Chao

    On the integer part of the

    -th root of a positive integer 143

    Yang Mingshun and Li Hailong

    On the additive cubic complements 147Liang Fangchi, Yi Yuan

    An arithmetical function and its hybrid mean value 151 Li Chao and Li Junzhuang

    On the -th power free sieve sequence 155Guo Jinbao and Zhao Xiqing

    On a new Smarandache sequence 159Zhao Xiaopeng and Yang Mingshun

    On some asymptotic formulae involving Smarandache multiplicative functions 163

    Li Junzhuang and Liu Duansen

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    8/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    9/178

    Preface

    Arithmetic is where numbers run across your mind looking for the answer.

    Arithmetic is like numbers spinning in your head faster and faster until you blow up with the

    answer.

    KABOOM!!

    Then you sit back down and begin the next problem.

    (Alexander Nathanson)

    Number theory is an ancient subject, but we still cannot answer many simplest and most naturalquestions about the integers. Some old problems have been solved, but more arise. All the research

    for these ancient or new problems implicated and are still promoting the development of numbertheory and mathematics.

    American-Romanian number theorist Florentin Smarandache introduced hundreds of interestsequences and arithmetical functions, and presented many problems and conjectures in his life. In

    1991, he published a book named Only problems, Not solutions!. He presented 105 unsolvedarithmetical problems and conjectures about these functions and sequences in it. Already many

    researchers studied these sequences and functions from his book, and obtained important results.

    This book,Research on Smarandache Problems in Number Theory (Collected papers), contains 41research papers involving the Smarandache sequences, functions, or problems and conjectures on

    them.All these papers are original. Some of them treat the mean value or hybrid mean value of

    Smarandache type functions, like the famous Smarandache function, Smarandache ceil function, orSmarandache primitive function. Others treat the mean value of some famous number theoretic

    functions acting on the Smarandache sequences, like k-th root sequence, k-th complement sequence,or factorial part sequence, etc. There are papers that study the convergent property of some infinite

    series involving the Smarandache type sequences. Some of these sequences have been first

    investigated too. In addition, new sequences as additive complement sequences are first studied inseveral papers of this book.

    Most authors of these papers are my students. After this chance, I hope they will be more interestedin the mysterious integer and number theory!

    All the papers are supported by the N. S. F. of P. R. China (10271093). So I would like to thank the

    Department of Mathematical and Physical Sciences of N. S. F. C.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    10/178

    I would also like to thank my students Xu Zhefeng and Zhang Xiaobeng for their careful typeset

    and design works. My special gratitude is due to all contributors of this book for their great help tothe publication of their papers and their detailed comments and corrections.

    More future papers by my students will focus on the Smarandache notions, such as sequences,functions, constants, numbers, continued fractions, infinite products, series, etc. in number theory!

    August 10, 2004

    Zhang Wenpeng

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    11/178

    AN ARITHMETIC FUNCTION AND THE PRIMITIVE

    NUMBER OF POWER

    Zhang WenpengResearch Center for Basic Science, Xian Jiaotong University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract For any fixed prime

    , we define

    "

    $ &

    ( ) 2

    $ 4

    6 8 (

    ) "

    E &

    ) 2

    E I Q 4

    T

    The main purpose of this paper is to study the mean value properties of)

    ,

    and give an interesting asymptotic formula for it.

    Keywords: Primitive number; Mean value; Asymptotic formula.

    W

    1. Introduction

    Let be a prime,X

    be any positive integer, we define two arithmetic func-

    tions as following:

    Y ` b

    X c e g i p q

    r t

    g u v X x

    r

    b

    g c u i

    v

    ` b

    X c e g i p q

    t

    X v X x

    X

    In problem 49 and 68 of reference [1], Professor F.Smarandache asked us to

    study the properties of these two arithmetic functions. About these problems,

    many scholars showed great interests in them (See references [2], [3]). But it

    seems that no one knows the relationship between these two arithmetic func-tions before. In this paper, we shall use the elementary methods to study the

    mean value properties ofv

    ` b Y ` b

    X c c, and give an interesting asymptotic for-

    mula for it. That is , we shall prove the following conclusion:

    Theorem. For any fixed prime and any real number , we have the

    asymptotic formula

    r

    v

    ` b Y ` b

    X c c e

    b

    c j

    l n

    Taking e

    ,

    in the theorem, we may immediately obtain the following

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    12/178

    2 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Corollary. For any real number

    , we have the asymptotic formula

    r

    v

    j

    b Y

    j

    b

    X c c e

    m l n

    i

    r

    v

    b Y

    b

    X c c e

    m l n

    W

    2. One simple lemma

    To complete the proof of the theorem, we need the following simple lemma:

    Lemma. For any fixed prime and real number

    , we have

    |

    }

    j

    |

    e

    j

    b

    c

    l ~

    j

    Proof. First we come to calculate

    e

    |

    r

    }

    j

    |

    Note that the identities

    e

    r

    |

    }

    j

    |

    r

    |

    }

    j

    |

    e

    X

    j

    r

    r

    |

    b

    }

    c

    j

    }

    j

    |

    e

    X

    j

    r

    r

    |

    }

    |

    i

    and

    j

    e

    ~

    X

    j

    r

    r

    |

    }

    |

    e

    j

    X

    j

    X

    j

    r

    j

    r

    |

    }

    |

    r

    |

    j

    }

    |

    e

    j

    X

    j

    X

    j

    r

    j

    r

    |

    j

    |

    X

    j

    X

    j

    b

    X c

    r

    j

    e

    b

    r

    c

    r

    r

    X

    j

    b

    X

    j

    X c

    r

    j

    So we have

    e

    ~

    b

    r

    c

    r

    r

    X

    j

    b

    X

    j

    X c

    r

    j

    n

    `

    j

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    13/178

    An arithmetic function and the primitive number of power 3

    e

    b

    c j

    b

    r

    c

    r

    j

    b

    c

    X

    j

    b

    X

    j

    X c

    r

    b

    c j

    Then we can immediately obtain

    |

    }

    j

    |

    e

    b

    c

    j

    b

    c

    l~

    j

    e

    j

    b

    c

    l ~

    j

    This completes the proof of the Lemma.W

    3. Proof of the Theorem

    In this section, we shall use the above Lemma to complete the proof of the

    Theorem. From the definition ofY

    `b

    X c

    andv

    `b

    X c

    , we may immediately get

    r

    v

    ` b Y ` b

    X c c

    e

    `

    }

    j

    e

    `

    `

    }

    j

    e

    `

    }

    j

    `

    `

    e

    `

    }

    j

    `

    `

    b

    x c

    e

    `

    }

    j

    `

    b

    x c

    `

    e

    `

    }

    j

    `

    `

    I Q

    e

    `

    }

    j

    |

    |

    l

    b

    c

    e

    |

    `

    }

    j

    |

    |

    `

    }

    j

    |

    l

    |

    `

    }

    j

    e

    |

    `

    }

    j

    |

    l n

    e

    ~

    j

    b

    c

    l ~

    j

    l n

    e

    b

    c j

    l n

    This completes the proof of the Theorem.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    14/178

    4 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Liu Hongyan and Zhang Wenpeng. A number theoretic function and its

    mean value property. Smarandache Notions Journal, 2002, 13: 155-159.

    [3] Zhang Wenpeng and Liu Duansen. On the primitive number of power

    and its asymptotic property. Smarandache Notions Journal, 2002, 13: 173-175.

    [4] Apostol T M. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    15/178

    ON THE PRIMITIVE NUMBERS OF POWER

    AND

    -POWER ROOTS

    Yi YuanResearch Center for Basic Science, Xian Jiaotong University, Xian, Shaanxi, P.R.China

    [email protected]

    Liang FangchiSchool of Science, Air Force Engineering University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract Let be a prime,

    be any positive integer,

    denotes the smallest integer

    I

    , where

    $

    &

    (

    . In this paper, we study the mean value properties

    of

    )

    $

    , where)

    $

    is the superior integer part of -power roots, and give an

    interesting asymptotic formula for it.

    Keywords: Primitive numbers of power ; -power roots; Asymptotic formula.

    W

    1. Introduction and results

    Let be a prime,X

    be any positive integer,Y ` b

    X cdenotes the smallest

    integer such thatY ` b

    X c uis divisible by r . For example,

    Y

    b

    c e ,

    Y

    b

    c e

    ,Y

    b

    c e ,Y

    b -

    c e , . In problem 49 of book [1], Professor F.

    Smarandache ask us to study the properties of the sequence

    Y ` b

    X c . About

    this problem, Professor Zhang and Liu in [2] have studied it and obtained an

    interesting asymptotic formula. That is, for any fixed prime and any positive

    integerX

    ,

    Y ` b

    X c e

    b

    c X

    l

    X

    For any fixed positive intger

    , letv

    r

    denotes the superior integer part of

    -

    power roots, that is,v

    e ,

    ,

    v

    j

    e ,

    v

    j

    e ,

    . In problem 80 of book

    [1], Professor F. Smarandache ask us to study the properties of the sequence

    v

    r

    . About this problem, the author of [3] have studied it and obtained an

    interesting asymptotic formula. That is, for any real number

    ,

    r

    b

    v

    r

    c e

    b

    c

    l

    i

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    16/178

    6 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    where

    b

    X cdenotes the numbers of all prime divisor of

    X,

    be a computable

    constant.

    In this paper, we will use the elementary method to study the asymptotic

    properties ofY ` b

    v

    r

    cin the following form:

    r

    t

    Y`

    b

    v

    r

    c

    Y`

    b

    v

    r

    c

    t

    i

    where

    be a positive real number, and give an interesting asymptotic formula

    for it. In fact, we shall prove the following result:

    Theorem. For any real number , let be a prime and X be any

    positive integer. Then we have the asymptotic formula

    r

    t

    Y ` b

    v

    r

    c

    Y ` b

    v

    r

    c

    t

    e

    Q

    l

    i

    where l denotes the l -constant depending only on parameter

    .

    W

    2. Proof of the Theorem

    In this section, we shall complete the proof of the theorem. First we need

    following one simple Lemma. That is,

    Lemma. Let be a prime andX

    be any positive integer, then we have

    t

    Y ` b

    X

    c

    Y ` b

    X c

    t

    e

    i if r t t g u

    iotherwise

    i

    whereY ` b

    X c e g , r g u denotes that r t g u and r g u .

    Proof. Now we will discuss it in two cases.b

    p c LetY ` b

    X c e g , if r g u , then we have r t g u and r g u . From the

    definition ofY

    `b

    X cwe have r

    b

    g

    c u, r

    b

    g

    c u,

    , r

    b

    g

    c uand r t

    b

    g

    c u, so

    Y ` b

    X

    c e g

    , then we get

    t

    Y`

    b

    X

    c

    Y`

    b

    X c

    t

    e

    (1)

    b

    p p c

    Let

    Y ` b

    X c e g

    , if

    r

    t

    g u

    and

    r

    t

    g u

    , then we have

    Y ` b

    X

    c e g

    ,so

    t

    Y ` b

    X

    c

    Y ` b

    X c

    t

    e

    (2)

    Combining (1) and (2), we can easily get

    t

    Y ` b

    X

    c

    Y ` b

    X c

    t

    e

    iif r

    g u

    i otherwise

    This completes the proof of Lemma.

    Now we use above Lemma to complete the proof of Theorem. For any real

    number

    , let

    be a fixed positive integer such that

    b

    c

    ,

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    17/178

    On the primitive numbers of power

    and -power roots 7

    then from the definition ofY ` b

    X cand the Lemma we have

    r

    t

    Y ` b

    v

    r

    c

    Y ` b

    v

    r

    c

    t (3)

    e

    r

    t

    Y ` b

    v

    r

    c

    Y ` b

    v

    r

    c

    t

    r

    t

    Y ` b

    v

    r

    c

    Y ` b

    v

    r

    c

    t

    e

    t

    Y ` b

    c

    Y ` b

    c

    t

    r

    t

    Y ` b

    v

    r

    c

    Y ` b

    v

    r

    c

    t

    e

    t

    Y ` b

    c

    Y ` b

    c

    t

    e

    Q

    `

    l

    b

    c (4)

    whereY ` b

    c e g. Note that if

    g u, then we have (see reference [4],

    Theorem 1.7.2)

    e

    g

    e

    g

    e g

    l n

    `

    g

    e

    g

    l

    g

    (5)

    From (4), we can deduce that

    g e

    b

    c

    l

    (6)

    So that

    g

    b

    c

    Q

    l

    i if

    Q

    Note that for any fixed positive integer

    , if there has oneg

    such that

    g u,

    then

    b

    g

    c u,

    b

    g

    c u,

    ,

    b

    g

    c u. Hence there have

    times ofg

    such that

    e

    `

    in the interval

    g

    b

    c

    Q

    l

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    18/178

    8 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Then from this and (3), we have

    r

    t

    Y ` b

    v

    r

    c

    Y ` b

    v

    r

    c

    t

    e

    Q

    `

    l

    b

    c

    e

    b

    c

    Q

    l

    l

    b

    c

    e

    Q

    l

    This completes the proof of Theorem.

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Zhang Wenpeng and Liu Duansen. primitive numbers of power and its

    asymptotic property, Smaramche Notions Journal 2002, 13: 173-175.

    [3] Yang Hai. Yanan University masters degree dissertion, 2004, 19-23.

    [4] Pan Chengdong and Pan Chengbiao, The Elementary number Theory,

    Beijing University Press Beijing, 2003.[5] Apostol T M. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    19/178

    MEAN VALUE ON THE PSEUDO-SMARANDACHE

    SQUAREFREE FUNCTION

    Liu HuaningDepartment of Mathematics, Northwest University, Xian, Shaanxi, P.R.China

    [email protected]

    Gao JingSchool of Science, Xian Jiaotong University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract For any positive integer

    , the pseudo-Smarandache squarefree function

    is defined as the least positive integer

    such that

    $

    is divisible by

    . In this

    paper, we study the mean value of

    , and give a few asymptotic formulae.

    Keywords: Pseudo-Smarandache squarefree function; Mean value; Asymptotic formula.

    W

    1. Introduction

    According to [1], the pseudo-Smarandache squarefree function b

    X c is

    defined as the least positive integerg

    such thatg

    r is divisible byX

    . It is

    obvious that

    b

    c e . For

    X , Maohua Le [1] obtained that

    b

    X c e

    j

    i(1)

    where

    , j

    ,

    , are distinct prime divisors ofX

    . Also he showed that

    r

    b

    b

    X c c

    i v i v

    is divergence.

    In this paper, we study the mean value of

    b

    X c, and give a few asymptotic

    formulae. That is, we shall prove the following:

    Theorem 1. For any real numbers }i

    with

    }

    and }

    , we have

    r

    |

    b

    X c

    X

    e

    b

    c

    b

    }

    c

    b

    }

    c

    `

    |

    i

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    20/178

    10 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    where

    b

    c is the Riemann zeta function,

    `

    denotes the product over all prime

    numbers.

    Theorem 2. For any real numbers }

    and

    , we have

    r

    |

    b

    X c e

    b

    }

    c

    |

    b

    c

    b

    }

    c

    `

    |

    b

    c

    ln

    |

    Q

    Noting that

    r

    b

    X c e

    l

    b

    cand

    |

    I

    }

    b

    }

    c e , so from

    Theorem 2 we immediately have the limit

    |

    I

    }

    `

    |

    b

    c

    e

    b

    c

    W

    2. Proof of the theorems

    Now we prove the theorems. For any real numbers } ,

    with

    }

    and

    }

    , let

    q

    b

    c e

    r

    |

    b

    X c

    X

    From (1) and the Euler product formula [2] we have

    q

    b

    c e

    `

    |

    |

    j

    e

    `

    `

    `

    e

    `

    ~

    `

    `

    |

    e

    b

    c

    b

    }

    c

    b

    }

    c

    `

    |

    This proves Theorem 1.

    For any real numbers }

    and

    , it is obvious that

    t

    |

    b

    X c

    t

    X

    |

    and

    r

    |

    b

    X c

    X

    }

    i

    where is the real part of

    . So by Perron formula [3] we can get

    r

    |b

    X c

    X

    e

    p

    q

    b

    c

    d

    l ~

    b

    c

    l

    b

    c

    i

    l

    b

    c

    i

    t t

    t t

    i

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    21/178

    Mean value on the pseudo-Smarandache squarefree function 11

    where

    is the nearest integer to

    , and t t

    t t

    e

    t

    t . Taking

    e

    ,

    e

    }

    j

    and

    in the above, then we have

    r

    |

    b

    X c e

    p

    |

    |

    q

    b

    c

    d

    l~

    |

    Now we move the integral line from } j

    p

    to } j

    p

    . This time,

    the function

    q

    b

    c

    have a simple pole point at e } with residue

    b

    }

    c

    |

    b

    c

    b

    }

    c

    `

    |

    b

    c

    Now taking

    e , then we have

    r

    |

    b

    X c e

    b

    }

    c

    |

    b

    c

    b

    }

    c

    `

    |

    b

    c

    p

    |

    Q

    |

    Q

    q

    b

    c

    d

    l n

    |

    Q

    e

    b

    }

    c

    |

    b

    c

    b

    }

    c

    `

    |

    b

    c

    l~

    q

    }

    p

    |

    Q

    b

    t

    t

    c

    d

    ln

    |

    Q

    e

    b

    }

    c

    |

    b

    c

    b

    }

    c

    `

    |

    b

    c

    l n

    |

    Q

    This completes the proof of Theorem 2.

    References

    [1] Maohua Le. On the pseudo-Smarandache squarefree function. Smaran-dache Notions Journal, 2002, 13: 229-236.

    [2] Apostol T M. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

    [3] Chengdong Pan and Chengbiao Pan. Foundation of Analytic Number

    Theory. Beijing: Science Press, 1997.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    22/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    23/178

    ON THE ADDITIVE

    -TH POWER COMPLEMENTS

    Xu ZhefengDepartment of Mthematics, Northwest University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract In this paper, similar to the Smarandache

    -th power complements, we defined

    the additive -th power complements. Using the elementary method, we study

    the mean value properties of the additive square complements, and give some

    interesting asymptotic formulae.

    Keywords: Additive

    -th power complements; Mean value; Asymptotic formula.

    W

    1. Introduction

    For any positive integerX

    , the Smarandache

    -th power complements

    b

    X c

    is the smallest positive integer such thatX

    b

    X cis a complete

    -th power, see

    problem 29 of [1]. Similar to the Smarandache

    -th power complements, wedefine the additive

    -th power complementsv

    b

    X cas follows:

    v

    b

    X cis the

    smallest nonnegative integer such thatv

    b

    X c

    Xis a complete

    -th power.

    For example, if e

    , we have the additive square complements sequence

    v

    j

    b

    X c

    b

    X e i i cas follows:

    i i i

    i

    -

    i i i i

    i

    i i

    -

    i i i i

    i i .

    In this paper, we stdudy the mean value properties ofv

    b

    X cand

    x

    b

    v

    b

    X c c,

    where xb

    X c is the Dirichlet divisor function, and give several interesting asymp-

    totic formulae. That is, we shall prove the following conclusion:

    Theorem 1. For any real number

    , we have the asymptotic formula

    r

    v

    b

    X c e

    j

    -

    j

    Q

    ln

    j

    Theorem 2. For any real number

    , we have the asymptotic formula

    r

    x

    b

    v

    b

    X c c e

    l n

    Q

    i

    where is the Euler constant.

    W

    2. Some lemmas

    Before the proof of the theorems, some lemmas will be usefull.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    24/178

    14 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Lemma 1. For any real number

    , we have the asymptotic formula:

    r

    x

    b

    c e

    b

    c

    l n

    Q

    i

    where is the Euler constant.

    Proof. See reference [2].

    Lemma 2. For any real number and any nonnegative arithmetical

    function qb

    X c with qb

    c e

    , we have the asymptotic formula:

    r

    q

    b

    v

    b

    X c c e

    Q

    r

    q

    b

    X c

    l

    r

    n

    Q

    q

    b

    X c

    !

    !

    !

    i

    where # $

    denotes the greatest integer less than or equal to

    and %b

    c e

    p

    .

    Proof. For any real number

    , let

    be a fixed positive integer such

    that

    b

    c

    Noting that ifX

    pass through the integers in the interval

    i

    b

    c

    , then

    v

    b

    X c pass through the integers in the inteval

    i

    b

    c

    and qb

    c e

    , we can write

    r

    q

    b

    v

    b

    X c c e

    r &

    q

    b

    v

    b

    X c c

    r

    q

    b

    v

    b

    X c c

    e

    r

    q

    b

    X c

    r &

    q

    b

    X c i

    where %b

    c e

    p

    . Since e

    Q

    , so we have

    r

    q

    b

    v

    b

    X c c e

    Q

    r

    q

    b

    X c

    l

    r

    n

    Q

    q

    b

    X c

    !

    !

    !

    This proves Lemma 2.

    Note: This Lemma is very usefull. Because if we have the mean value for-

    mula ofq

    b

    X c, then from this lemma, we can easily get the mean value formula

    of

    r

    q

    b

    v

    b

    X c c.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    25/178

    On the additive -th power complements 15

    W

    3. Proof of the theorems

    In this section, we will complete the proof of the theorems. First we prove

    Theorem 1. From Lemma 1 and the Euler summation formula (See [3]), let

    q

    b

    X c e X, we have

    r

    v

    b

    X c e

    Q

    r

    X

    l

    r

    n

    Q

    X

    !

    !

    !

    e

    Q

    j

    j

    j

    l n

    j

    e

    j

    -

    j

    Q

    l n

    j

    This proves Theorem 1.

    Now we prove Theorem 2. From Lemma 1 and Lemma 2, we have

    r

    x

    b

    v

    b

    X c c

    e

    Q

    r

    x

    b

    X c

    l

    r

    n

    Q

    x

    b

    X c

    !

    !

    !

    e

    Q

    l

    b

    c

    l n

    Q

    e

    Q

    n

    b

    c

    b

    c

    l

    b

    j

    c

    l n

    Q

    e

    b

    c

    Q

    b

    c

    Q

    l n

    Q

    Then from the Euler summation formula, we can easily get

    r

    x

    b

    v

    b

    X c c e

    l n

    Q

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    26/178

    16 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    This completes the proof of the theorems.

    References

    [1] F.Smaradache. Only problems, not solutions, Xiquan Publishing House,

    Chicago, 1993.

    [2] G.L.Dirichlet. Sur lusage des series infinies dans la theorie des nom-

    bres. Crelles Journal, No.18, 1938.

    [3] Tom M. Apostol. Introduction to Analytic Number Theory, Springer-

    Verlag, New York, 1976, pp54.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    27/178

    ON THE SMARANDACHE PSEUDO-MULTIPLES

    OF ( SEQUENCE

    Wang Xiaoying1. Research Center for Basic Science, Xian Jiaotong University, Xian, Shaanxi, P.R.China

    2. Department of Mathematics, Northwest University, Xian, Shaanxi, [email protected]

    Abstract The main purpose of this paper is to study the mean value properties of the

    Smarandache pseudo-multiples of ) number sequence, and give an interesting

    asymptotic formula for it.

    Keywords: Pseudo-multiples of ) numbers; Mean value; Asymptotic formula.

    W

    1. Introduction

    A number is a pseudo-multiple of

    if some permutation of its digits is a

    multiple of , including the identity permutation. For example:

    i i

    i i

    i i

    i i

    -

    i

    i i i are pseudo-multiple of

    numbers. Let

    de-

    notes the set of all the pseudo-multiple of

    numbers. In reference [1], Profes-

    sor F. Smarandache asked us to study the properties of the pseudo-multiple of

    sequence. About this problems, it seems that none had studied it, at least we

    have not seen such a paper before. In this paper, we use the elementary method

    to study the mean value properties of this sequence, and obtain an interesting

    asymptotic formula for it. That is, we shall prove the following:

    Theorem. For any real number , we have the asymptotic formula

    r 0 1

    r

    q

    b

    X c e

    r

    q

    b

    X c

    l n

    3 5 6

    3 5

    Q

    i

    where e 7 9

    r

    t

    q

    b

    X c

    t

    . Taking qb

    Xd c e x

    b

    X c,

    b

    X cas the Dirichlet

    divisor function and the function of the number of prime factors respectively,

    then we have the following:

    Corollary 1. For any real number

    , we have the asymptotic formula

    r 0 1

    r

    x

    b

    X c e

    b

    c

    l n

    3 5 6

    3 5

    Q

    i

    where is the Euler constant,

    is any fixed positive number.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    28/178

    18 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Corollary 2. For any real number

    , we have the asymptotic formula

    r 0 1

    r

    b

    X c e

    l

    i

    where

    is a computable constant.

    W

    2. Proof of the Theorem

    Now we completes the proof of the Theorem. First let

    b

    c , then

    . According to the definition of set

    , we

    know that the largest number of digits (

    ) not attribute set

    isC

    . Infact, in these numbers, there are C one digit, they are

    i i i

    -

    i

    i i C i ; There

    are C j two digits; The number of the

    digits are C

    . So the largest number of

    digits (

    ) not attribute set

    is C C j

    C

    e DF

    b

    C

    c

    C

    . Since

    C

    C

    e

    n

    C

    6

    Q

    3 G H

    6

    Q

    e

    b

    c

    Q

    3 G H

    6

    Q

    e

    3 5 6

    3 5

    Q

    So we have,

    C

    e

    l n

    3 5 6

    3 5

    Q

    Next, let

    denotes the upper bounds of tq

    b

    X c

    t

    b

    X

    c, then

    r 0

    1

    r

    q

    b

    X c e

    l n

    3 5 6

    3 5

    Q

    Finally, we have

    r 0 1

    r

    q

    b

    X c e

    r

    q

    b

    X c

    r 0

    1

    r

    q

    b

    X c

    e

    r

    q

    b

    X c

    ln

    3 5 6

    3 5

    Q

    This proves the Theorem.

    Now the Corollary 1 follows from the Theorem, the asymptotic formula

    r

    x

    b

    X c e

    b

    c

    l n

    Q

    (see [2]), and the estimate xb

    X c P

    (for all

    X

    ). And then, the

    Corollary 2 follows from the Theorem, the asymptotic formula

    r

    b

    X c e

    l

    (See [3]), and the estimate

    b

    X c P

    (for all

    X

    ).

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    29/178

    On the Smarandache pseudo-multiples of sequence 19

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Tom M. Apostol. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

    [3] G.H. Hardy and S. Ramanujan. The normal number of prime factors of

    a numberX

    . Quart. J. Math. 48(1917), 76-92.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    30/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    31/178

    AN ARITHMETIC FUNCTION AND THE DIVISOR

    PRODUCT SEQUENCES

    Zhang TianpingDepartment of Mathematics, Northwest University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract Let

    be any positive integer, R S

    denotes the product of all positive divisors

    of

    . Let

    be a prime,)

    denotes the largest exponent (of power

    ) such

    that divisible by

    . In this paper, we study the asymptotic properties of the mean

    value of)

    RS

    , and give an interesting asymptotic formula for it.

    Keywords: Divisor product sequences; Mean value; Asymptotic formula.

    W

    1. Introduction

    LetX

    be any positive integer, U b

    X cdenotes the product of all positive

    divisors of X . That is, U b

    X c e

    `

    x . For example, U b

    c e i U

    b

    c e

    i U

    b

    c e i U

    b -

    c e C i Let be a prime,

    v

    ` b

    X cdenotes the largest ex-

    ponent (of power ) such that

    r

    t

    X. In problem 25 and 68 of reference [1],

    Professor F.Smarandache asked us to study the properties of these two arith-

    metic functions. About these problems, many scholars showed great interests

    in them (see references [2],[3]). But it seems that no one knows the relation-

    ship between these two arithmetic functions before. In this paper, we shall

    use the elementary methods to study the mean value properties ofv

    `b Y

    `b

    X c c,

    and give an interesting asymptotic formula for it. That is , we shall prove the

    following conclusion:

    Theorem. Let

    be a prime, then for any real number

    , we have theasymptotic formula

    r

    v

    ` b

    U

    b

    X c c e

    b

    c

    b

    c

    b

    c X

    `

    -

    j

    b

    b

    c

    `

    l

    b

    Q

    c i

    where is the Euler constant, and

    denotes any fixed positive number.

    Taking e

    ,

    in the theorem, we may immediately obtain the following

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    32/178

    22 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Corollary. For any real number

    , we have the asymptotic formula

    r

    v

    j

    b

    U

    b

    X c c e

    l

    b

    Q

    c

    r

    v

    b

    U

    b

    X c c e

    C

    -

    -

    l

    b

    Q

    c

    W

    2. Some lemmas

    To complete the proof of the theorem, we need the following simple lem-

    mas:

    Lemma 1. For any positive integerX

    , we have the identity

    U

    b

    X c e X

    Se

    $ f

    i

    wherex

    b

    X cis the divisor function.

    Proof. This formula can be immediately got from Lemma 1 of [2].

    Lemma 2. For any real number , we have the asymptotic formula

    r

    r

    x

    b

    X c e

    `

    `

    j

    l

    b

    Q

    c i

    where

    `

    denotes the product over all primes, is the Euler constant, and

    denotes any fixed positive number.

    Proof. Let

    e

    j

    i

    b

    c e

    `

    j

    Then by the Perron formula

    (See Theorem 2 of reference [4]), we may obtain

    r

    r

    x

    b

    X c e

    p

    j

    b

    c

    b

    c

    x

    l

    b

    Q

    c i

    where

    b

    c is the Riemann-zeta function.

    Moving the integral line from j

    p

    to j

    p

    . This time, the function

    q

    b

    c e

    j

    b

    c

    b

    c

    has a second order pole point at e with residue

    `

    `

    j

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    33/178

    An arithmetic function and the divisor product sequences 23

    So we have

    p

    ~

    Q

    Q

    Q

    Q

    j

    b

    c

    b

    c

    x

    e

    `

    `

    j

    Note that

    p

    ~

    Q

    Q

    Q

    Q

    j

    b

    c

    b

    c

    x

    P

    Q

    From the above we can immediately get the asymptotic formula:

    r

    r

    x

    b

    X c e

    `

    `

    j

    l

    b

    Q

    c

    This completes the proof of Lemma 2.

    Lemma 3. Let be a prime, then for any real number , we have the

    following asymptotic formulae

    |

    }

    |

    e

    b

    c j

    l

    i(1)

    |

    }

    j

    |

    e

    j

    b

    c

    l~

    j

    i(2)

    |

    }

    |

    e

    -

    j

    b

    c

    `

    l~

    (3)

    Proof. We only prove formulab

    cand

    b

    c. First we come to calculate

    q e

    |

    r

    }

    j

    |

    Note that the identities

    q

    e

    r

    |

    }

    j

    |

    r

    |

    }

    j

    |

    e

    X

    j

    r

    r

    |

    b

    }

    c

    j

    }

    j

    |

    e

    X

    j

    r

    r

    |

    }

    |

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    34/178

    24 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    and

    q

    j

    e

    ~

    X

    j

    r

    r

    |

    }

    |

    f

    e

    j

    X

    j

    X

    j

    r

    j

    r

    |

    }

    |

    r

    |

    j

    }

    |

    e

    j

    X

    j

    X

    j

    r

    j

    r

    |

    j

    |

    X

    j

    X

    j

    b

    X c

    r

    j

    e

    b

    r

    c

    r

    r

    X

    j

    b

    X

    j

    X c

    r

    j

    So we have

    q e

    ~

    b

    r

    c

    r

    r

    X

    j

    b

    X

    j

    X c

    r

    j

    n

    `

    j

    e

    b

    c j

    b

    r

    c

    r

    j

    b

    c

    X

    j

    b

    X

    j

    X c

    r

    b

    c j

    Then we can immediately obtain

    |

    }

    j

    |

    e

    b

    c j

    b

    c

    l ~

    j

    e

    j

    b

    c

    l~

    j

    This proves formulab

    c.

    Now we come to prove formulab

    c. Let

    %e

    |

    r

    }

    |

    Note that the identities

    %

    e

    r

    |

    }

    |

    r

    |

    }

    |

    e

    X

    r

    r

    |

    b

    }

    c

    }

    |

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    35/178

    An arithmetic function and the divisor product sequences 25

    e

    X

    r

    r

    |

    j

    }

    j

    }

    |

    e

    X

    r

    r

    r

    r

    ~

    j

    X

    r

    r

    j

    r

    r

    `

    ~

    -

    j

    X

    j

    r

    ~

    X

    r

    b

    r

    c

    r

    r

    `

    `

    e

    j

    -

    b

    c j

    X X

    j

    r

    b

    X c

    r

    b

    c

    X

    r

    b

    r

    c

    b

    r

    j

    c

    b

    c

    r

    b

    c

    Then we have

    |

    }

    |

    e

    ~

    j

    -

    b

    cj

    b

    c

    b

    c

    l ~

    e

    -

    j

    b

    c

    `

    l~

    This completes the proof of Lemma 3.

    W

    3. Proof of the Theorem

    In this section, we shall use the above lemmas to complete the proof of the

    Theorem. From the definition of U b

    X cand

    v

    `b

    X c, we may immediately get

    r

    v

    `b

    U

    b

    X c c

    e

    ` g

    `

    g

    b

    }

    c

    }

    x

    b i

    c e

    `

    b

    }

    c

    }

    g

    `

    `

    g

    x

    b i

    c

    e

    |

    `

    b

    }

    c

    }

    ~

    |

    |

    j

    ln

    Q

    e

    j

    |

    `

    b

    }

    c

    }

    |

    |

    `

    b

    }

    c

    }

    j

    |

    l n

    Q

    e

    j

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    36/178

    26 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    p

    ~

    j

    b

    c

    b

    c j

    l~

    j

    ~

    -

    j

    b

    c

    `

    j

    b

    c

    l ~

    ln

    Q

    e

    b

    c

    b

    c

    b

    c X

    `

    -

    j

    b

    b

    c

    `

    l n

    Q

    This completes the proof of the Theorem.

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Liu Hongyan and Zhang Wenpeng. On the divisor products and proper

    divisor products sequences. Smarandache Notions Journal, 2002, 13: 128-133.

    [3] Liu Hongyan and Zhang Wenpeng. A number theoretic function and its

    mean value property. Smarandache Notions Journal, 2002, 13: 155-159.

    [4] Pan Chengdong and Pan Chengbiao. Elements of the analytic number

    Theory. Beijing: Science Press, 1991.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    37/178

    THE SMARANDACHE IRRATIONAL ROOT SIEVE

    SEQUENCES

    Zhang XiaobengDepartment of Mathematics, Northwest University, Xian, Shaanxi, P.R.China

    [email protected]

    Lou YuanbingCollege of Science, Tibet University, Lhasa, Tibet, P.R.China

    [email protected]

    Abstract In this paper, we use the analytic method to study the mean value properties of

    the irrational root sieve sequence, and give an interesting asymptotic formula for

    it.

    Keywords: Smarandache irrational root sieve; Mean value; Asymptotic formula.

    W

    1. Introduction

    According to reference [1], the definition of Smarandache irrational root

    sieve is: from the set of natural numbers (except

    and

    ):

    -take off all powers of

    ,

    ;

    -take off all powers of

    ,

    ;

    -take off all powers of

    , ;

    -take off all powers of

    ,

    ;

    -take off all powers of

    ,

    ;

    -take off all powers of

    , ;

    and so on (take off all

    -powers,

    ). For example:

    i i i

    i i

    i

    i i i

    -

    i i i C i are all irrational root sieve sequence. Let

    denotes the set of all the irrational root sieve. In reference [1], Professor F.

    Smarandache asked us to study the properties of the irrational root sieve se-

    quence. About this problem, it seems that none had studied it, at least we have

    not seen such a paper before. In this paper, we study the mean value of the

    irrational root sieve sequence, and give an interesting asymptotic formula for

    it. That is, we shall prove the following:

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    38/178

    28 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Theorem. Let xb

    X c denote the divisor function. Then for any real number

    , we have the asymptotic formula

    r 0 1

    r

    x

    b

    X c

    e

    -

    j

    q

    Q

    j

    j

    Q

    Q

    `

    q

    b

    c

    r

    q

    t

    Q

    ln

    Q

    u

    v

    u

    i

    where

    denotes any fixed positive number, is the Euler constant,

    i

    j

    i

    i

    `

    i

    r

    i

    t

    are the computable constants.

    W

    2. Some Lemmas

    To complete the proof of the theorem, we need the following lemmas:

    Lemma 1. For any real number , we have the asymptotic formula:

    r

    x

    b

    X c e

    b

    c

    l n

    Q

    u

    v

    u

    i

    where

    denotes any fixed positive number and is the Euler constant.

    Proof. This result may be immediately got from [2].

    Lemma 2. For any real number , we have two asymptotic formulae

    r x

    x

    b

    X

    j

    c e

    q

    j

    -

    j

    q

    j

    q

    l n

    Q

    v

    r

    Q

    x

    b

    X

    c e

    Q

    `

    Q

    j

    j

    Q

    Q

    l n

    Q

    i

    where

    i

    j

    i

    i

    i

    j

    i

    are computable constants.

    Proof. Let

    q

    b

    c e

    r

    x

    b

    X

    j

    c

    X

    i

    Reb

    c . Then from the Euler product formula [3] and the multiplicative

    property ofx

    b

    X cwe have

    q

    b

    c e

    `

    j

    e

    `

    j

    e

    `

    ~

    j

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    39/178

    The Smarandache irrational root sieve sequences 29

    e

    `

    j

    e

    b

    c

    b

    c

    where

    b

    c is the Riemann zeta-function. By Perron formula [2] with

    e

    ,

    e

    Q

    and

    e

    j

    , we have

    r

    x

    b

    X

    j

    c e

    p

    b

    c

    b

    c

    x

    ln

    Q

    To estimate the main term

    p

    b

    c

    b

    c

    x i

    we move the integral line from e

    j

    p

    to e

    j

    p

    . This time, the

    function

    q

    b

    c e

    b

    c

    b

    c

    has a three order pole point at e

    with residue

    u

    ~

    b

    c

    b

    c

    b

    c

    j

    e

    j

    j

    j

    i

    where

    i

    j

    are the computable constants.

    Note that

    p

    ~

    Q

    Q

    Q

    Q

    b

    c

    b

    c

    x P

    Q

    From above we may immediately get the asymptotic formula:

    r

    x

    b

    X

    j

    c e

    j

    j

    j

    ln

    Q

    That is,

    r x

    x

    b

    X

    j

    c e

    q

    j

    -

    j

    q

    j

    q

    l n

    Q

    v

    This proves the first formula of Lemma 2.

    Similarly, we can deduce the second asymptotic formula of Lemma 2. In

    fact let

    %

    b

    c e

    r

    x

    b

    X

    c

    X

    i

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    40/178

    30 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Reb

    c . Then from the Euler product formula [3] and the multiplicative

    property ofx

    b

    X cwe have

    %

    b

    c e

    `

    -

    j

    e

    `

    j

    e

    `

    ~

    j

    e

    `

    j

    e

    `

    b

    c

    j

    b

    c

    `

    b

    c j

    where

    b

    cis the Riemann zeta-function. Then by Perron formula [2] and the

    method of proving the first asymptotic formula of Lemma 2 we may immedi-

    ately get

    r

    x

    b

    X

    c e

    `

    j

    j

    l n

    Q

    That is,

    r

    Q

    x

    b

    X

    c e

    Q

    `

    Q

    j

    j

    Q

    Q

    l n

    Q

    i

    This proves the Lemma 2.

    W

    3. Proof of the Theorem

    Now we completes the proof of the Theorem. According to the definition of

    the set

    and the result of Lemma 1 and Lemma 2 , we have

    r 0 1

    r

    x

    b

    X c

    e

    r

    x

    b

    X c

    r x

    x

    b

    X

    j

    c

    r

    Q

    x

    b

    X

    c

    l

    `

    3 5

    3 5

    r

    Q

    x

    b

    X

    c

    !

    e

    r

    x

    b

    X c

    r x

    x

    b

    X

    j

    c

    r

    Q

    x

    b

    X

    c

    l

    `

    3 5

    3 5

    Q

    !

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    41/178

    The Smarandache irrational root sieve sequences 31

    e

    ~

    q

    -

    j

    Q

    j

    j

    Q

    Q

    `

    q

    b

    c

    r

    q

    t

    Q

    ln

    Q

    u

    v

    u

    i

    where

    (p e i i i

    ) are computable constants.

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Pan Chengdong and Pan Chengbiao. Elements of the analytic numberTheory. Beijing: Science Press, 1991.

    [3] Tom M. Apostol. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    42/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    43/178

    A NUMBER THEORETIC FUNCTION AND ITS MEAN

    VALUE

    Lv ChuanDepartment of Mathematics, Northwest University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract Let

    be a prime,

    denote the largest exponent of power

    which divides

    . In this paper, we study the properties of this sequence

    , and give an

    interesting asymptotic formula for

    $

    E

    .

    Keywords: Asymptotic formula; Largest exponent; Mean value.

    W

    1. Introduction

    Let be a prime, ` b

    X c denote the largest exponent of power which di-

    videsX

    . In problem 68 of [1], Professor F.Smarandach asked us to study the

    properties of the sequence `

    b

    X c. This problem is closely related to the factor-

    ization of X u . In this paper, we use elementary methods to study the asymptotic

    properties of the mean value

    r

    `

    b

    X c, and give an interesting asymptotic

    formula for it. That is, we will prove the following:

    Theorem. Let be a prime, g

    be an integer. Then for any real number

    , we have the asymptotic formula

    r

    `

    b

    X c e

    v

    ` b

    g c

    l

    b

    c i

    where v` b

    g c is a computable constant.

    Taking g e i i in the theorem, we may immediately obtain the follow-ing

    Corollary. For any real number , we have the asymptotic formula

    r

    ` b

    X c e

    l

    b

    j

    c

    r

    j

    `

    b

    X c e

    b

    c j

    l

    b

    c

    r

    `

    b

    X c e

    j

    -

    b

    c

    l

    b

    `

    c

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    44/178

    34 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    W

    2. Proof of the Theorem

    In this section, we complete the proof of the theorem. In fact, from the

    definition of ` b

    X cwe have

    r

    `

    b

    X c e

    `

    `

    `

    }

    e

    |

    3 G H

    3 G H

    }

    `

    e

    |

    3 G H

    3 G H

    }

    |

    l

    b

    c

    e

    |

    3 G H

    3 G H

    }

    |

    l

    |

    3 G H

    3 G H

    }

    !

    Let

    v

    ` b

    g c e

    r

    X

    r

    i

    then v` b

    g c is a computable constant. Obviously we have

    |

    3 G H

    3 G H

    }

    |

    e

    r

    X

    r

    |

    3 G H

    3 G H

    }

    |

    e v

    ` b

    g c

    l

    3 G H

    3 G H

    n

    `

    r

    e v

    `b

    g c

    l~

    ~

    e v

    ` b

    g c

    l n

    (1)

    and

    l

    |

    3 G H

    }

    !

    e

    ln

    (2)

    From (1) and (2) we have

    r

    `

    b

    X c e

    n

    v

    ` b

    g c

    l n

    l n

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    45/178

    A number theoretic function and its mean value 35

    e

    v

    ` b

    g c

    0 l n

    This completes the proof of the theorem. As to v` b

    g c , it is easy to show that

    v

    `b

    c e

    r

    r

    e

    i

    and

    v

    ` b

    g c e

    r

    b

    X

    c

    r

    e

    r

    X

    r

    r

    X

    r

    r

    X

    r

    r

    r

    e v

    ` b

    g c

    v

    ` b

    g c

    v

    ` b

    c

    v

    ` b

    c i

    so we have

    v

    ` b

    g c e

    n

    v

    ` b

    g c

    v

    ` b

    c

    v

    ` b

    c

    From this formula, we can easily compute the first severalv

    ` b

    g c:

    v

    ` b

    c e

    b

    c j

    i v

    ` b

    c e

    j

    b

    c

    i v

    ` b

    c e

    -

    j

    b

    c

    `

    i

    Then use the Theorem, we can get the Corollary.

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Liu Hongyan and Zhang Wenpeng. A number theoretic function and its

    mean value property. Smarandache Notions Journal, 2002, 13: 155-159.

    [3] Apostol T M. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    46/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    47/178

    ON THE PRIMITIVE NUMBERS OF POWER

    AND

    ITS TRIANGLE INEQUALITY

    Ding LipingDepartment of Mathematics, Northwest University, Xian, Shaanxi, P.R.China

    dingding [email protected]

    Abstract The main purpose of this paper is using the elementary method to study the

    properties of

    , and give a triangle inequality for it.

    Keywords: Primitive numbers; Arithmetical property; triangle inequality.

    W

    1. Introduction

    Let be a prime, X be any fixed positive integer,Y ` b

    X c denote the smallest

    positive integer such thatY ` b

    X c uis divisible by r . For example,

    Y

    b

    c e ,

    Y

    b

    c e

    ,Y

    b

    c e

    ,Y

    b -

    c e

    ,Y

    b

    c e

    ,

    . In problem 49 of book

    [1], Professor F. Smarandache asks us to study the properties of the sequenceY ` b

    X c. About this problem, some asymptotic properties of this sequence have

    been studied by many scholar. In this paper, we use the elementary methods to

    study the arithmetical properties ofY ` b

    X c, and give a triangle inequality for it.

    That is, we shall prove the following:

    Theorem 1. Let be an odd prime, g be positive integer. Then we have

    the triangle inequality

    Y `

    ~

    g

    Y ` b

    g

    c

    Theorem 2. There are infinite integersg

    b

    p e i i i csatisfying

    Y `

    ~

    g

    e

    Y ` b

    g

    c

    W

    2. Proof of the theorems

    In this section, we complete the proof of the theorems. First we prove the-

    orem 1. From the definition ofY ` b

    X c, we know that

    t

    Y ` b

    g

    c u, t

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    48/178

    38 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Y ` b

    g c u(

    p e ). From this we can easily obtain:

    e

    t

    Y ` b

    g

    c u

    Y ` b

    g c u

    t

    b Y ` b

    g

    c

    Y ` b

    g c c u (1)

    But from the definition ofY ` b

    X c, we know that

    Y ` b

    X c uis the smallest positive

    integer that is divisible by r . That is

    s

    t

    Y`

    b

    g

    g

    c u

    (2)

    From (1), (2) we can immediately getY ` b

    g

    g c

    Y ` b

    g

    c

    Y ` b

    g c

    Now the theorem 1 follows from this inequality and the induction.

    Next we complete the proof of theorem 2. For any positive integersg

    withg

    e g

    (

    p i

    ), we let }

    e

    }

    b

    i X csatisfy

    |

    X u. Then

    }

    e

    }

    b

    i X c e

    X

    For convenient, we let

    e

    `

    E

    `

    . Since

    e

    s

    j

    e

    e

    So we haveY

    `b

    c e

    i p e i i i (3)

    On the other hand,

    jk

    k

    k

    k

    k

    l

    mn

    n

    n

    n

    n

    e

    e

    So

    Y `

    ~

    e

    (4)

    Combining (3) and (4) we may immediately obtain

    Y `

    ~

    e

    Y ` b

    c

    This completes the proof of Theorem 2.

    References

    [1] Jozsef Sandor, On an generalization of the Smarandache function, Note

    Numb.Th.Diser.Math. 5(1999),41-51

    [2] Tom M A. Introduction to Analytic Number Theory. New York, 1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    49/178

    THE ADDITIVE ANALOGUE OF SMARANDACHE

    SIMPLE FUNCTION

    Zhu Minhui1. Math and Phys Dept, XAUEST, Xian, Shaanxi, P.R.China

    2. Department of Mathematics, Northwest University, Xian, Shaanxi, [email protected]

    Abstract The main purpose of this paper is to study the asymptotic properties of

    ,

    and give two interesting asymptotic formulae for it.

    Keywords: Smarandache-simple function; Additive Analogue; Asymptotic formula.

    W

    1. Introduction and results

    For any positiveX

    , the Smarandache functionY b

    X c

    is defined as the smallest

    g

    , where X t g u . For a fixed prime , the Smarandache-simple functionY ` b

    X c is defined as the smallest g

    , where

    r

    t

    g u . In reference [2],Jozsef Sandor introduced the additive analogue of the Smarandache-simple

    functionY ` b

    cas follows:

    Y`

    b

    c e

    g

    g u i

    b

    i c i

    andY

    `

    b

    c e

    7 9

    g

    g u

    i # i c i

    which is defined on a subset of real numbers. It is clear thatY ` b

    c e gif

    b b

    g c u i g u$ for

    g (for

    g e it is not defined, as 0!=1! ), therefore

    this function is defined for

    . About the properties ofY b

    X c

    , many people

    had studied it before (See [2], [3]). But for the asymptotic formula ofY ` b

    c , it

    seems that no one have studied it before. The main purpose of this paper is tostudy the asymptotic properties of

    Y ` b

    c, and obtain an interesting asymptotic

    formula for it. That is, we shall prove the following:

    Theorem 1. For any real number , we have the asymptotic formula

    Y ` b

    c e

    l

    j

    Obviously, we have

    Y ` b

    c e

    Y

    `

    b

    c

    i p q

    b

    g u i

    b

    g

    c u c

    b

    g c

    Y

    `

    b

    c i p q e

    b

    g

    c u

    b

    g c

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    50/178

    40 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Therefore, we can easily get the following:

    Theorem 2. For any real number , we have the asymptotic formula

    Y

    `

    b

    c e

    l

    j

    W

    2. Proof of the theorem

    In this section, we complete the proof of the theorem 1. In fact, from the

    definition ofY ` b

    c, we have

    b

    g c u

    g uand taking the logistic

    computation in the two sides of the inequality, we get

    p

    p (1)

    Then using the Eulers summation formula we have

    p e

    x

    b

    #

    $c

    b

    c }

    x

    e g

    g g

    l

    b

    g c (2)

    and

    p e

    x

    b

    #

    $c

    b

    c x

    e g

    g g

    l

    b

    g c (3)

    Combining (1),(2) and (3), we can easily deduce that

    e g

    g g

    l

    b

    g c (4)

    So

    g e

    g

    l

    b

    c (5)

    Similarly, we continue taking logistic computation in two sides of (5), then we

    also have

    g e

    l

    b

    g c (6)

    and

    g e

    l

    b

    c (7)

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    51/178

    The additive analogue of Smarandache simple function 41

    Hence, by (5), (6) and (7) we have

    Y ` b

    c e

    l

    b

    g c

    l

    b

    c

    e

    l

    b

    g c

    b

    l

    b

    g c c

    e

    l

    j

    This completes the proof of Theorem 1.

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Jozsef Sandor. On additive analogue of certain arithmetic functions,

    Smaramche Notes Journal 2004, 14: 128-132.

    [3] Mark Farris and Patrick Mitchell, Bounding the smarandache functoin,

    Smaramche Notes Journal 2002, 13: 37-42

    [4] Tom M.Apostol, Introduction to Analytic Number Theory, Springer-

    Verlag, New York,1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    52/178

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    53/178

    ON THE

    -POWER COMPLEMENT SEQUENCE

    Yao WeiliResearch Center for Basic Science, Xian Jiaotong University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract The main purpose of this paper is using analytic method to study the asymp-

    totic properties of -power complement sequence, and give several interesting

    asymptotic formulae.

    Keywords: -power complement sequence; asymptotic formula; mean value.

    W

    1. Introdution

    For any positive integerX

    , let

    b

    X cdenotes

    -power complement

    sequence. That is,

    b

    X c denotes the smallest integer such that X

    b

    X c be a

    perfect

    -power. In problem 29 of reference [1], professor F.Smarandache

    asked us to study the properties of this sequence. About this problem, somepeople had studied it before, see references [4]and [5]. The main purpose

    of this paper is using the analytic method to study the asymptotic properties

    of

    -power complement sequence, and obtain several interesting asymptotic

    formulae. That is, we shall prove the following :

    Theorem. Let xb

    X c denote the Dirichlet divisor function, then for any real

    number

    , we have the asymptotic formula

    r

    x

    b

    X

    b

    X c c e

    b

    c

    l

    b

    Q

    c i

    where

    ,

    ,

    ,

    are computable constants, is any fixed positive num-

    ber.

    From this theorem, we may immediately deduce the following

    Corollary 1. Letv

    b

    X cbe the square complement sequence, then for any

    real number

    , we have

    r

    x

    b

    X v

    b

    X c c e

    b

    j

    c

    l

    b

    Q

    c i

    where

    ,

    ,

    are computable constants.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    54/178

    44 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Corollary 2. Let

    b

    X c be the cubic complement sequence, then for any real

    number , we have

    r

    x

    b

    X

    b

    X c c e

    b

    j

    c

    l

    b

    Q

    c i

    where

    ,

    ,

    and

    are computable constants.

    W

    2. Proof of the Theorem

    In this section, we shall complete the proof of the Theorem. Let

    q

    b

    c e

    r

    x

    b

    X

    b

    X c c

    X

    From the definition of

    b

    X c, the properties of the divisor function and the

    Euler product formula [2], we have

    q

    b

    c e

    `

    ~

    x

    b

    b

    c c

    x

    b

    j

    b

    j

    c c

    j

    e

    `

    ~

    x

    b

    c

    x

    b

    c

    x

    b

    j

    c

    x

    b

    j

    c

    j

    e

    `

    j

    e

    `

    `

    p

    p

    `

    `

    p

    `

    e

    b

    c

    `

    b

    c

    e

    b

    c

    b

    c

    `

    b

    c

    b

    c

    j

    n

    b

    c

    i

    where

    b

    cis the Riemann Zeta function. Obviously, we have

    t

    x

    b

    X

    b

    X c c

    t

    X i

    r

    x

    b

    X

    b

    X c c

    X

    i

    where is the real part of

    . Therefore by Perron formula [3]

    r

    x

    b

    X

    b

    X c c

    X

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    55/178

    On the -power complement sequence 45

    e

    p

    q

    b

    c

    x

    l~

    b

    c

    l

    b

    c

    i

    l

    b

    c

    i

    i

    where

    be the nearest integer to

    ,

    e

    t

    t . Taking

    e

    ,

    e

    j

    ,

    e ,

    b

    c e ,

    b

    c e

    , then we have

    r

    x

    b

    X

    b

    X c c e

    p

    b

    c

    b

    c

    b

    c

    x

    l

    b

    Q

    c i

    where

    b

    c e

    `

    b

    c

    b

    c

    j

    n

    b

    c

    To estimate the main term

    p

    b

    c

    b

    c

    b

    c

    x i

    we move the integer line from e

    p

    to e

    p

    , then the function

    b

    c

    b

    c

    b

    c

    have one order pole point at e with residue

    u

    b

    c

    b

    c

    b

    c

    b

    c

    e

    u

    n

    n

    b

    c

    b

    c

    b

    c

    b

    c

    n

    n

    b

    c

    b

    c

    b

    c

    b

    c

    n

    b

    c

    b

    c

    b

    c

    b

    c

    e

    b

    c i

    where

    i

    i i

    are computable constants. So we can obtain

    p

    ~

    Q

    Q

    Q

    Q

    b

    c

    b

    c

    b

    c x

    e

    b

    c

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    56/178

    46 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Note that the estimates

    p

    Q

    b

    c

    b

    c

    b

    c x

    P

    Q

    i

    p

    Q

    b

    c

    b

    c

    b

    c x

    P

    Q

    i

    and

    p

    Q

    Q

    b

    c

    b

    c

    b

    c x

    P

    Q

    Therefore we get

    r

    x

    b

    X

    b

    X c c e

    b

    c

    l

    b

    Q

    c

    This completes the proof of the Theorem .

    References

    [1]. F. Smarandache. Only Problems, Not solutions. Chicago: Xiquan

    Publishing House, 1993.

    [2]. Apostol T M. Introduction to Analytic Number Theory. New York:

    Springer- Verlag, 1976.[3]. Pan Chengdong and Pan Chengbiao. Foundation of Analytic Number

    Theory. Beijing: Science press, 1997.

    [4]. Zhu Weiyi. On the -power complement and -power free number

    sequence. Smarandache Notions Journal, 2004, 14: 66-69.

    [5]. Liu Hongyan and Lou Yuanbing. A Note on the 29-th Smarandaches

    problem. Smarandache Notions Journal, 2004, 14: 156-158.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    57/178

    ON THE INFERIOR AND SUPERIOR FACTORIAL

    PART SEQUENCES

    Li JieDepartment of Mathematics, Northwest University, Xian, Shaanxi, P.R.China

    [email protected]

    Abstract The main purpose of this paper is using the elementary method to study the con-

    vergent property of an infinite series involving the inferior and superior factorial

    part sequences, and give an sufficient condition of the convergent property of the

    series.

    Keywords: Inferior factorial part; Superior factorial part; Infinite series.

    W

    1. Introduction

    For any positive integerX

    , the inferior factorial part denoted byv

    b

    X cis the

    largest factorial less than or equal to X . It is the sequence: i i i i i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i . On the other

    hand, the superior factorial part denoted by

    b

    X cis the smallest factorial greater

    than or equal toX

    . It is the sequence as follows: i i

    i

    i

    i

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    -

    i

    i

    i

    i In the

    42th problem (see [1]) of his famous book l Xi

    i

    g i

    i

    p

    X ,

    Professor F.Smarandache asked us to study these sequences. About this prob-

    lem, it seems no one had studied it before. In this paper, we studied two infinite

    series involving vb

    X c and

    b

    X c as follows:

    e

    r

    v

    |

    b

    X c

    i

    Y

    e

    r

    |

    b

    X c

    i

    and given some sufficient conditions of the convergent property of them. That

    is, we shall prove the following

    Theorem. Let } be any positive real number. Then the infinite series

    andY

    are convergent if }

    , divergent if }

    .

    Especially, when }e

    , we have the following

    Corollary. We have the identity

    r

    v j

    b

    X c

    e

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    58/178

    48 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    W

    2. Proof of the theorem

    In this section, we will complete the proof of the theorem. Letv

    b

    X c e g u. It

    is easy to see that if g u

    X

    b

    g

    c u , then vb

    X c e g u . So the same number

    g urepeated

    b

    g

    c u g utimes in the sequence

    v

    b

    X c

    b

    X e i i c.

    Hence, we can write

    e

    r

    v

    |

    b

    X c

    e

    b

    g

    c u g u

    b

    g u c

    |

    e

    g g u

    b

    g u c

    |

    e

    g

    b

    g u c

    |

    It is clear that if }

    then

    is convergent, if }

    then

    is divergent. Using

    the same method, we can also get the sufficient condition of the convergent

    property ofY

    . Especially, when }e

    , from the knowledge of mathematical

    analysis (see [2]), we have

    r

    v j

    b

    X c

    e

    b

    g c u

    e

    g u

    e

    This completes the proof of the theorem.

    References

    [1] F.Smarandache. Only problems, Not solutions, XiQuan Publishing House,

    Chicago, 1993, pp36.

    [2] T.M.Apostol. Mathematical analysis, 2nd ed., Addison Wesley Publish-

    ing Co., Reading, Mass, 1974.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    59/178

    A NUMBER THEORETIC FUNCTION AND ITS MEAN

    VALUE

    Gao NanSchool of Sciences, Department of Mathematics, Xian Shiyou University, Xian, Shaanxi, P.R.China

    gaonan 0 0 [email protected]

    Abstract Let

    and are primes,

    denotes the largest exponent of power which

    divides

    . In this paper, we study the properties of this sequence e

    $ f

    , and

    give an interesting asymptotic formula for

    $

    e

    $ f

    .

    Keywords: Asymptotic formula; Largest exponent; Mean value.

    W

    1. Introduction

    Let and

    are two primes,

    b

    X c

    denotes the largest exponent of power

    which dividesX

    . It is obvious that e

    if

    dividesX

    but

    does not.

    In problem 68 of [1], Professor F.Smarandache asked us to study the propertiesof the sequence

    b

    X c. In this paper, we use elementary methods to study the

    asymptotic properties of the mean value

    r

    r , where is a prime such

    that , and give an interesting asymptotic formula for it. That is, we will

    prove the following :

    Theorem. Let and are primes with . Then for any real number

    , we have the asymptotic formula

    r

    r

    e

    `

    l n

    Q

    i

    if

    `

    `

    `

    n

    `

    `

    `

    b

    c

    `

    j

    `

    l n

    Q

    i if e

    where

    is any fixed positive number, is the Euler constant.

    W

    2. Proof of the Theorem

    In this section, we will complete the proof of the theorem. Let

    q

    b

    c e

    r

    r

    X

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    60/178

    50 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    For any positive integerX

    , it is clear that

    b

    X cis a addtive function. So

    r

    is a multiplicative function. Then from the definition of

    b

    X cand the Euler

    product formula (See Theorem 11.6 of [3]), we can write

    q

    b

    c e

    r

    r

    X

    e

    `

    Q

    ~

    `

    E

    Q

    e

    `

    Q

    ~

    `

    Q

    `

    Q

    j

    e

    ~

    j

    j

    `

    Q

    j

    e

    b

    c

    By Perron formula (See reference [2]), let

    e

    ,

    e ,

    e

    j . Then we

    have

    r

    r

    e

    p

    j

    j

    b

    c

    b

    c

    x

    l

    b

    j

    c i

    where

    b

    c e

    `

    and is any fixed positive number.

    Now we estimate the main term

    p

    j

    j

    b

    c

    b

    c

    x i

    we move the integral line from

    p

    to

    p

    .

    If

    , then function

    b

    c

    b

    c

    have a simple pole point at e

    , so we have

    p

    ~

    j

    j

    j

    j

    j

    j

    j

    j

    b

    c

    b

    c

    x e

    b

    c

    Taking

    e

    , we have

    p

    ~

    Q

    j

    j

    Q

    b

    c

    b

    c

    x

    P

    j

    Q

    b

    p

    c

    b

    c

    j

    x

    P

    j

    e

    Q

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    61/178

    A number theoretic function and its mean value 51

    And we can easy get the estimate

    p

    Q

    Q

    b

    c

    b

    c

    x

    P

    b

    p

    c

    b

    c

    Q

    x

    P

    Q

    Note that

    b

    c e

    i

    we may immediately obtain the asymptotic formula

    r

    r

    e

    l

    b

    Q

    c i

    if

    .

    If e

    , then the function

    b

    c

    b

    c

    have a second order pole point at e

    . Let Res n

    b

    c

    b

    c

    denotes its

    residue, so we have

    b

    c

    b

    c

    e

    b

    c

    b

    c

    j

    e

    ~

    b

    c

    b

    c

    b

    c

    b

    c

    b

    b

    c

    b

    c c

    Noting that

    b

    c

    e

    i

    b

    c

    b

    c e

    and

    b

    b

    c

    b

    c c e i

    we may immediately get

    b

    c

    b

    c

    e

    b

    c

    So we have the asymptotic formula

    r

    r

    e

    b

    c

    b

    c

    l

    b

    Q

    c i

    if e

    . This completes the proof of Theorem.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    62/178

    52 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    References

    [1] Smarandache F. Only problems, not Solutions. Chicago: Xiquan Publ.

    House, 1993.

    [2] Pan Chengdong and Pan Chengbiao. Elements of the analytic number

    Theory. Beijing: Science Press, 1991, pp 98.

    [3] Apostol T M. Introduction to Analytic Number Theory. New York:

    Springer-Verlag, 1976.

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    63/178

    ON THE GENERALIZED CONSTRUCTIVE SET

    Gou suDepartment of Mathematics and Physics, Xian Institute of Posts and Telecoms, Xian, Shaanxi,

    P.R.China

    [email protected]

    Abstract In this paper, we use the elementary methods to study the convergent properties

    of the series

    I

    $ Q

    8

    )

    $

    where)

    $

    is a generalized constructive number, and is any positive number.

    Keywords: Generalized constructive set; Series; Convergent properties.

    W

    1. Introduction

    The generalized constructive setY

    is defined as: numbers formed by digits

    x

    i x

    j

    i i x

    only, allx

    being different each other,

    g

    . That is to

    say,

    (1) x

    i x

    j

    i i x

    belong toY

    ;

    (2) Ifv i

    belong toY

    , thenv

    belongs toY

    too;

    (3) Only elements obtained by rules (1) and (2) applied a finite number of

    times belong toY

    .

    For example, the constructive set (of digits 1, 2) is : i i i i i i i

    i i i i i i i i i i And the constructive

    set (of digits 1, 2, 3) is : i i i i i i i i i i i i i i i

    i i i i i i i i i In problem 6, 7 and 8 of ref-

    erence [1], Professor F.Smarandache asked us to study the properties of this

    sequence. About this problem, it seems that no one had studied it before. For

    convenience, we denote this sequence by v

    r

    . In this paper, we shall use the

    elementary methods to study the convergent properties of the series

    r

    v

    |

    r

    i

    where } is any positive number. That is , we shall prove the following conclu-

    sion:

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    64/178

    54 RESEARCH ON SMARANDACHE PROBLEMS IN NUMBER THEORY

    Theorem 1. The series

    r

    v

    |

    r

    is convergent if } g , and divergent

    if}

    g .

    Especially, let

    Y

    j

    e

    i

    and

    Y

    e

    Then we have the following

    Theorem 2. The seriesY

    j

    andY

    are convergent, and the estimate

    -

    Y

    j

    C

    -

    v X x

    -

    C

    Y

    -

    - - -

    hold.

    W

    2. Proof of the Theorems

    In this section, we shall complete the proof of the Theorems. First we prove

    Theorem 1. Note that for e i i i i

    there areg

    numbers of

    digits in

    the generalized constructive sequence, so we have

    r

    v

    |

    r

    g

    b

    c

    |

    e g

    g

    |

    e

    g

    e

    g

    |

    |

    g

    i

    where we have used the fact that the series

    g

    |

    is convergent only if its common proportion

    ithat is }

    g. This

    completes the proof of the Theorem 1.

    Now we come to prove Theorem 2.

    Y

    j

    e

  • 8/14/2019 Research on Smarandache Problems in Number Theory, edited by Zhang Wenpeng

    65/178

    On the generalized constructive set 5