How many lattice points lie on the sphere x 2 + y 2 + z 2 = 2 0 1 3 ?
Details and assumptions
A lattice point has integer coordinates.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
We will first assume that x , y , z are positive, and not care about the order.
Working modulo 4, since 2 0 1 3 ≡ 1 ( m o d 4 ) , and the quadratic residues modulo 4 are 0 and 1, it follows that exactly 1 of the integers is odd. Working modulo 16, 2 0 1 3 ≡ 1 3 ( m o d 1 6 ) , and the quadratic residues modulo 16 are 0, 1, 4, 9, we can only get 1 3 = 0 + 4 + 9 . Hence, the numbers must be of the form 4 A , 8 B ± 2 , 8 C ± 3 .
We know that the largest possible integer is 2 0 1 3 < 4 5 . Let's check for possible values of C = 3 , 5 , 1 1 , 1 3 , 1 9 , 2 1 , 2 7 , 2 9 , 3 5 , 3 7 , 4 3 . Recall Fermat's 2 square theorem, which tells us that a number can be written in the form a 2 + b 2 , if and only if every prime of the form 4 k + 3 appears an even number of times in the expansion.
2
0
1
3
−
3
2
=
2
0
0
4
=
4
×
3
×
1
6
7
will not work.
2
0
1
3
−
5
2
=
4
×
7
×
7
1
will not work.
2
0
1
3
−
1
1
2
=
4
×
1
1
×
4
3
will not work.
2
0
1
3
−
1
3
2
=
4
×
4
6
1
yields the solution
(
3
8
,
2
0
,
1
3
)
.
2
0
1
3
−
1
9
2
=
4
×
7
×
5
9
will not work.
2
0
1
3
−
2
1
2
=
4
×
3
×
1
3
1
will not work.
2
0
1
3
−
2
7
2
=
4
×
3
×
1
0
7
will not work.
2
0
1
3
−
2
9
2
=
4
×
2
9
3
yields the solution
(
3
4
,
4
,
2
9
)
.
2
0
1
3
−
3
5
2
=
4
×
1
9
7
yields the solution
(
2
8
,
2
,
3
5
)
.
2
0
1
3
−
3
7
2
=
4
×
7
×
2
3
will not work.
2
0
1
3
−
4
3
2
=
4
×
4
1
yields the solution
(
1
0
,
8
,
4
3
)
.
Hence, there are 4 triples of positive numbers that work. Since the numbers are distinct, and we care about the order, there are 4 × 6 = 2 4 triples of positive numbers. Since each sign could be positive or negative, there are 2 4 × 8 = 1 9 2 lattice points on the sphere.
"quadratic residues modulo 4 are 0 and 1" whats meant by this,sir?
Would you please help me to solve the following question,
x 2 + y 2 ≤ 2 5
How many ordered pairs ( x , y ) are there , where x and y are integers?
For easier manipulation we first consider a,b,c to be all non-negative integers.
First we note that squares of even integers give 0 (mod 4) and squares of odd integers give 1 (mod 4). 2013 is congruent to 1 mod 4, so we know that only one of a,b,c is even. Without loss of generalization, let a=2x, b=2y and c=2z+1.
We can rewrite the equation as (2x)^2 + (2y)^2 + (2z+1)^2 = 4x^2 + 4y^2 + 4z(z+1) + 1 = 2013. Subtracting 1 from both sides and dividing them by 4, we get x^2 + y^2 + z(z+1) = 503.
Now note that z(z+1) can only be either 0 mod 4 (when z is 0 or 3 mod 4) or 2 mod 4 (when z is 1 or 2 mod 4). Since 503 is 1 mod 4, the only possibility is that x and y are both odd, and z is 1 or 2 mod 4. We thus split into 2 cases.
If z is 1 mod 4, we let x=2m, y=2n and z=4k+1. Expanding the equation we get 4m^2 + 4n^2 + 4n + 1 + 16k^2 + 12k + 2 = 503, and after subtracting 3 on both sides and dividing by 4 we get m^2 + n(n+1) + k(4k+3) = 125.
If z is 2 mod 4, we let x=2m, y=2n and z=4k+2. Expanding the equation we get 4m^2 + 4n^2 + 4n + 1 + 16k^2 + 20k + 6 = 503, and after subtracting 7 on both sides and dividing by 4 we get m^2 + n(n+1) + k(4k+5) = 124.
So we get m^2 + n(n+1) = 125 - k(4k+3) or 124 - k(4k+5).
Since a,b,c are non-negative integers, m,n,k are non-negative integers as well. Therefore, k = 0 to 5 for the first case and 0 to 4 for the second. We split into cases where k = 0 to 5, where we can easily check for solutions.
When k = 5,
m^2 + n(n+1) = 10. This gives us a possible solution, when m=n=2.
When k = 4,
m^2 + n(n+1) = 40 or 49. This gives us a possible solution, when m=7, n=1.
When k = 3,
m^2 + n(n+1) = 73 or 80. This gives us a possible solution, when m=1, n =8.
When k = 2,
m^2 + n(n+1) = 98 or 103. There is no nonnegative integers m,n that satisfy the equation.
When k = 1,
m^2 + n(n+1) = 115 or 118. This gives us a possible solution, when m=5, n = 9.
When k = 0,
m^2 + n(n+1) = 124 or 125. There is no nonnegative integers m,n that satisfy the equation.
This gives us a total of 4 solutions for (m,n,k), where m,n,k are all positive. However, a,b,c are interchangeable, which gives us 4x6=24 solutions for (a,b,c) where all of a,b,c are positive.
Now note that any possible solution (p,q,r) would mean that the (± p, ± q, ± r) is a solution, and vice versa, so we have a total of 24x8=192 solutions in all.
Problem Loading...
Note Loading...
Set Loading...
Firstly, note that 2 0 1 3 ≡ 1 3 ( m o d 1 6 ) and that the only squares modulo 1 6 are 0 , 1 , 4 , 9 . The only way to get 1 3 from these is 0 + 4 + 9 . Suppose for now that x is the odd number, i.e. that x 2 ≡ 9 ( m o d 1 6 ) . Analysing which x satisfy this congruence, we deduce that x ≡ 3 , 5 ( m o d 8 ) . Before checking these x 's case by case, we eliminate a few more of them.
Note that 2 0 1 3 = 3 × 1 1 × 6 1 . In particular x 2 + y 2 + z 2 is divisible by 3 and 1 1 , but not by their squares. I claim that this implies that none of x , y , z can be divisible by either 3 or 1 1 , since both of these leave a remainder of 3 when divided by 4 .
To prove the claim, let p ≡ 3 ( m o d 4 ) be a prime, suppose that p divides x 2 + y 2 + z 2 , that p 2 does note divide it, and that p divides x . This means that p divides y 2 + z 2 , but that p 2 does not. However, since p ≡ 3 ( m o d 4 ) , if p divides y 2 + z 2 , then p divides both y and z , contradicting the fact that p 2 does not divide y 2 + z 2 .
We now check case by case the odd numbers x from 1 to 4 3 (which is the largest odd number less than 2 0 1 3 ) which are ≡ 3 , 5 ( m o d 8 ) and are not divisible by 3 or 1 1 . This leaves x ∈ { 5 , 1 3 , 1 9 , 2 9 , 3 5 , 3 7 , 4 3 } . For each of these 2 0 1 3 − x 2 = y 2 + z 2 is the sum of two even squares, and hence 4 2 0 1 3 − x 2 is also the sum of two squares. For those values of x , the corresponding values of 4 2 0 1 3 − x 2 ] are { 4 9 7 , 4 6 1 , 4 1 3 , 2 9 3 , 1 9 7 , 1 6 1 , 4 1 } . Among these we have 4 9 7 = 7 × 7 1 , 4 1 3 = 7 × 5 9 , 1 6 1 = 7 × 2 3 , while the other 4 are prime. We have seen that if 7 divides y 2 + z 2 , then 7 2 divides it, ruling out the first three options. On the other hand, the four primes are congruent to 1 modulo 4 , meaning that each is uniquely expressible as the sum of two squares (not counting permutation).
Hence there are 4 solutions where ( x , y , z have a certain ordering and are all positive. They can be rearranged in 3 ! = 6 ways, and minus signs can be introduced in 2 3 = 8 ways. This gives a total of 4 × 6 × 8 = 1 9 2 triples.