You can still use p-series. You can just change the indices on summation:
∑[n = 0..∞] 1/√(n+1)
Let n+1 = i
When n = 0, i = 1
∑[n = 0..∞] 1/√(n+1) = ∑[i = 1..∞] 1/√i
By p-series ∑[i = 1..∞] 1/√i diverges, since 0 < p ≤ 1
Therefore ∑[n = 0..∞] 1/√(n+1) diverges, since it is = ∑[i = 1..∞] 1/√i
------------------------------
Alternate method: p-series combined with comparison test:
For n ≥ 1
n + 1 ≤ 2n
√(n+1) ≤ √(2n)
1/√(n+1) ≥ 1/√(2n)
By p-series ∑[1..∞] 1/√(2n) = 1/√2 ∑[1..∞] 1/n^(1/2) diverges, since p = 1/2 ≤ 1
By comparison test, ∑[1..∞] 1/√(n+1) also diverges
If series actually starts from 0, then we can show that it also diverges, since the sum of a finite value and a diverging series is also diverging:
∑[0..∞] 1/√(n+1) = 1 + ∑[1..∞] 1/√(n+1)
------------------------------
Alternate method: Integral test:
∫[0 to ∞] 1/√(x+1) dx
= 2 √(x+1) |[0 to ∞]
= 2 { (lim[x→∞] √(x+1)) - √1 }
= 2 (∞ - 1)
= ∞
----> integral diverges
By integral test, series diverges
Mαthmφm