- 鐵幣
 - 9 元
 - 文章
 - 8 篇
 - 聲望
 - 10 枚
 - 上次登入
 - 09-8-21
 - 精華
 - 0
 - 註冊時間
 - 07-6-29
 - UID
 - 407111
  
 
 
 
            
 | 
    
        
        
高中 3 
I try another way 
Suppose for some p, q in Z+, 
     p/q = 1 + 1/2 + ... + 1/n, 
then 
     p = n!(1+1/2+...+1/n) and 
     q = n! 
p/q is an integer if and only if q divides p, i.e. 
     n! | n!(1+1/2+...+1/n). 
This is so if 
     n!(1+1/2+...+1/n) = n! + 1*3*...*n + ... + (n-1)! ≡ 0 (mod k) 
for k = 2, ..., n. 
However, there exists the largest prime, say p', in {1, 2, ..., n}, such that in modulo p', 
     n!(1+1/2+...+1/n) ≡ 1*2*...*(p'-1)(p'+1) ... n (mod p') 
which doesn't vanish, thus q does not divide p. |   
 
                                      
         
     
 |