Problem about the sum of the divisors of a number

  • Thread starter Thread starter phymath7
  • Start date Start date
  • Tags Tags
    Number theory Sum
Click For Summary
The discussion revolves around understanding the sum of divisors function, denoted as σ(n), and its relationship with classes of numbers defined by their prime factors. Participants clarify that for a number n to belong to the class N_k, it must satisfy the condition σ(n) ≥ k*n, indicating that the number of positive divisors is proportionally greater than k times the number itself. There is confusion regarding the definitions of k, N_k, and the notation used, particularly concerning the prime factors and their powers in the prime decomposition of n. An example is requested to illustrate these concepts, emphasizing the need for clarity in the definitions and conditions for the classes. The conversation highlights the mathematical intricacies involved in determining the properties of numbers based on their divisors.
phymath7
Messages
48
Reaction score
4
Homework Statement
The sum of divisor function σ(n) returns the sum of all positive divisors d of the number n. We denote ##N_k## any number that fulfils the following condition:
σ(##N_K##) ≥##k.N_K##.
Find examples for ##N_3##;##N_4##;##N_5## and prove that they fulfil this condition.
Relevant Equations
##σ(n)=\frac{p_{1}^{a_1+1}-1}{p_1-1}.\frac{p_{2}^{a_2+1}-1}{p_2-1}.......\frac{p_{k}^{a_k+1}-1}{p_k-1}##
I've found that ##N_1## is 1. But it's really tiresome to find them one by one. I also tried to use the equation but couldn't. Please help me out.
 
Physics news on Phys.org
I do not understand your condition. What is ##k##, what ##N_K##, what ##K##? Can you say what the condition for ##\sigma(N_3)## is?
 
Are the ##p_i## in your first equation the prime divisors of ##N##? Like fresh said, can you give as one worked example? Also, if '.' stands for product, please use \cdot instead? Edit: If you want a number that is less than the sum of its proper divisors, then 12, 18, 20 are examples. Is this what you're looking for?
 
Last edited:
fresh_42 said:
I do not understand your condition. What is ##k##, what ##N_K##, what ##K##? Can you say what the condition for ##\sigma(N_3)## is?
I believe the condition stated in words is that the number of positive divisors for ##N_k## is greater than or equal to ##k## times ##N_k##, for example, if ##n## is ##N_3## then it must have ##\sigma(n) \geq 3n##. So basically the ##N_k## are best thought of as classes.
 
GwtBc said:
I believe the condition stated in words is that the number of positive divisors for ##N_k## is greater than or equal to ##k## times ##N_k##, for example, if ##n## is ##N_3## then it must have ##\sigma(n) \geq 3n##. So basically the ##N_k## are best thought of as classes.
Yes, but what are the k's about? What is , e.g.##N_3##?
 
WWGD said:
Yes, but what are the k's about? What is , e.g.##N_3##?
If ##n## satisfies ##\sigma(n) \geq 3n## then ##n \in N_3##, so ##N_1 \subseteq N_2 \subseteq N_3 \subseteq ...##. OP denotes ##n## by ##N_3## instead but that would just be confusing since there is clearly more than one such number, and it doesn't make a difference to the problem so...

What's not clear to me is what the ##a_i## are. I assume the ##p_i## are the prime divisors of ##n##? @phymath7
 
The ##a_i## are the powers in the prime decomposition of ##n=p_1^{a_1}\cdots p_k^{a_k}##.

So for ##N_3## we have all numbers ##n## with three prime factors plus the condition that ##\sigma(n)\geq 3n##? ##\sigma(N_3)\geq 3N_3## still makes no sense: You defined it by using it! The closest I come from all what you wrote is
$$
N_3 = \{\,n\,|\,\exists \,p_1,p_2,p_3\text{ prime and }\exists\,a_1,a_2,a_3\in \mathbb{Z}\, : \,n=p_1^{a_1}p_2^{a_2}p_3^{a_3}\,\wedge\,\sigma(n)\geq 3n\,\,\}
$$
but that is a guess.
 
fresh_42 said:
The ##a_i## are the powers in the prime decomposition of ##n=p_1^{a_1}\cdots p_k^{a_k}##.

So for ##N_3## we have all numbers ##n## with three prime factors plus the condition that ##\sigma(n)\geq 3n##? ##\sigma(N_3)\geq 3N_3## still makes no sense: You defined it by using it! The closest I come from all what you wrote is
$$
N_3 = \{\,n\,|\,\exists \,p_1,p_2,p_3\text{ prime and }\exists\,a_1,a_2,a_3\in \mathbb{Z}\, : \,n=p_1^{a_1}p_2^{a_2}p_3^{a_3}\,\wedge\,\sigma(n)\geq 3n\,\,\}
$$
but that is a guess.
I'm sorry as there is some problem in the first post. The homework equation would be like this,if ##n=p_1^{a_1}\cdots p_j^{a_j}## then, ##σ(n) = \frac {p_1^{a_1+1}} {P_1-1}\cdots \frac {p_t^{a_j+1}} {P_j-1}##.I just mixed up with j and k.Now please do what you can.
 
So let's start with ##N_3##. We thus have
$$
3n = 3 \cdot p_1^{a_1}\cdots p_j^{a_j}\leq \dfrac{p_1^{a_{1}+1}-1}{p_1-1} \cdots \dfrac{p_j^{a_{j}+1}-1}{p_j-1}= \sigma(n)
$$
You are requested to give an example. So choose some primes and play with this inequality until you find ##p_i## and ##a_i## for which it holds.

There is also a trivial upper bound: a number ##n## cannot have more divisors than all numbers ##1,\ldots, \dfrac{n}{2}##. So ##\sigma(n) \leq 1+\ldots + \dfrac{n}{2}=\dfrac{1}{8}(n^2+2n)##. What does that mean for ##\sigma(n)\geq kn##?
 
Last edited: