I've Just Solved 1236 Problems on Brilliant

Yeah, I've just solved 1236 problems on Brilliant (as the Stats sub-section in the About section says)!

Long ago, I noticed a special property about the 4-digit number 1236.

The property is as follows:

The sum of its digits, 1 + 2 + 3 + 6 = 12 1+2+3+6=12 , is equal to the number 12 formed by the left half of the digits of 1236; the product of its digits, 1 × 2 × 3 × 6 = 36 1\times2\times3\times6=36 , is equal to the number 36 formed by the right half of the digits of 1236.

Let me precisely define the terms Left-Half and Right-Half for a positive integer with an even number of digits, where the leftmost digit is non-zero.

If P = L × 1 0 n + R P=L\times 10^n+R is a positive integer with 2 n ( > 0 ) 2n(>0) digits, then the number L L is the Left-Half of P P and the number R R is the Right-Half of P P .

To celebrate this moment, I've decided to share the following problem:

Find the sum of all positive integers P P with 2 n ( > 0 ) 2n(>0) digits and non-zero leftmost digit such that the sum of digits of P P is equal to its Left-Half and the product of digits of P P is equal to its Right-Half .


The answer is 1686.

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.

1 solution

Let A P A_P be the set of all positive integers P P with 2 n > 0 2n>0 digits and the leftmost digit is non-zero such that the sum of digits of P P is equal to its Left-Half and the product of digits of P P is equal to its Right-Half .

We also use S P S_P to denote the sum of digits of P P and G P G_P to denote the product of digits of P P .

We consider the decimal expansion of P = d 2 n 1 × 1 0 2 n 1 + d 2 n 2 × 1 0 2 n 2 + + d n × 1 0 n + + d 1 × 10 + d 0 P=d_{2n-1}\times10^{2n-1}+d_{2n-2}\times10^{2n-2}+ \ldots{ } \ldots{ } \ldots{ }+d_{n}\times10^{n}+\ldots{ } \ldots{ } \ldots{ }+d_1\times10+d_0 , with d 2 n 1 0 d_{2n-1}\neq0 .

Then L = d 2 n 1 × 1 0 n 1 + d 2 n 2 × 1 0 n 2 + + d n L=d_{2n-1}\times10^{n-1}+d_{2n-2}\times10^{n-2}+ \ldots{ } \ldots{ } \ldots{ }+d_{n} and R = d n 1 × 1 0 n 1 + d n 2 × 1 0 n 2 + + d 0 R=d_{n-1}\times10^{n-1}+d_{n-2}\times10^{n-2}+ \ldots{ } \ldots{ } \ldots{ }+d_{0} .

First let's prove such P P do not exist for n > 2 n>2 . To prove this, we'll show that L > S P L>S_P for n 3 n\geq 3 .

As d 2 n 1 0 d_{2n-1}\neq0 , we have L 1 0 n 1 L\geq 10^{n-1} . We also have 18 n S P 18n\geq S_P . So, it suffices to show that 1 0 n 1 > 18 n 10^{n-1}>18n for n 3 ( 1 ) n\geq3\ldots{ } \ldots{ } \ldots{ } (1) . We apply the Principle of Mathematical Induction on n n .

The inequality ( 1 ) (1) holds for n = 3 n=3 ; 1 0 3 1 = 1 0 2 = 100 > 54 = 18 × 3 10^{3-1}=10^2=100>54=18\times3 .

Assume that the inequality ( 1 ) (1) holds for n = m n=m ; that is, 1 0 m 1 > 18 m 10^{m-1}>18m .

Now, 1 0 m 1 > 18 m 10^{m-1}>18m

1 0 m 1 + 18 > 18 m + 18 \implies 10^{m-1}+18>18m+18 (Adding 18 18 to both sides)

1 0 m 1 + 9 × 1 0 m 1 > 1 0 m 1 + 18 > 18 m + 18 \implies 10^{m-1}+9\times 10^{m-1} >10^{m-1}+18 > 18m+18 (By inductive hypothesis)

10 × 1 0 m 1 > 18 ( m + 1 ) \implies 10\times10^{m-1}>18(m+1)

1 0 m > 18 ( m + 1 ) \implies 10^{m}>18(m+1)

1 0 ( m + 1 ) 1 > 18 ( m + 1 ) \implies 10^{(m+1)-1}>18(m+1) , that is, the inequality ( 1 ) (1) holds for n = m + 1 n=m+1 .

So, by PMI , 1 0 n 1 > 18 n 10^{n-1}>18n for n 3 n\geq3 . That means, such P P don't exist for n > 2 n>2 .


Let's see if there are such P P for n = 1 n=1 .

For n = 1 n=1 , the decimal expansion of P = d 1 × 10 + d 0 P=d_1\times 10+d_0 . Then, L = d 1 L=d_1 , R = d 0 R=d_0 .

P A P P\in A_P only if S P = d 1 + d 0 = L = d 1 d 1 + d 0 = d 1 d 0 = 0 S_P=d_1+d_0=L=d_1 \implies d_1+d_0=d_1 \implies d_0=0 .

So, d 0 = 0 d_0=0 is a necessary condition for P A P P\in A_P .

Moreover, it turns out that d 0 = 0 d_0=0 is also a sufficient condition for P A P P\in A_P . Because, if d 0 = 0 d_0=0 , then S P = d 1 + d 0 = d 1 = L S_P=d_1+d_0=d_1=L and G P = d 1 × d 0 = 0 = d 0 = R G_P=d_1\times d_0=0=d_0=R .

That means, P = d 1 × 10 + d 0 A P d 0 = 0 P=d_1\times 10+d_0 \in A_P \iff d_0=0 .

So, 10 , 20 , 30 , 40 , 50 , 60 , 70 , 80 , 90 A P \boxed{10,20,30,40,50,60,70,80, 90 \in A_P} .


Now let's see if there are such P P for n = 2 n=2 .

For n = 2 n=2 , the decimal expansion of P = d 3 × 1 0 3 + d 2 × 1 0 2 + d 1 × 10 + d 0 P=d_3\times 10^3+d_2\times10^2+d_1\times 10+d_0 . Then L = d 3 × 10 + d 2 L=d_3\times10+d_2 and R = d 1 × 10 + d 0 R=d_1\times10+d_0 .

Now, L = S P L=S_P

d 3 × 10 + d 2 = d 3 + d 2 + d 1 + d 0 \iff d_3\times10+d_2=d_3+d_2+d_1+d_0

9 × d 3 = d 1 + d 0 \iff 9\times d_3=d_1+d_0

As d 1 + d 0 18 d_1+d_0\leq18 and d 3 0 d_3\neq0 , there are two possibilities for d 3 d_3 , namely, 1 1 and 2 2 .

When d 3 = 2 d_3=2 , we have only one option for R = d 1 × 10 + d 0 R=d_1\times10+d_0 , namely R = 99 R=99 . Then P = 2 a 99 P=2a99 for some decimal digit a a . But P = 2 a 99 P=2a99 satisfies the G P = R G_P=R condition for NO a a . So, P = 2 a 99 ∉ A P P=2a99 \not \in A_P .

When d 3 = 1 d_3=1 , the options we have for R = d 1 × 10 + d 0 R=d_1\times10+d_0 are: 09 , 90 , 18 , 81 , 27 , 72 , 36 , 63 , 45 09, 90, 18, 81, 27, 72, 36, 63, 45 and 54 54 . Applying the same logic on each R R as we did in the case when d 3 = 2 d_3=2 , we'll able to find only one P P that belongs to A P A_P , namely, 1236 1236 .

Hance, A P = { 10 , 20 , 30 , 40 , 50 , 60 , 70 , 80 , 90 , 1236 } \boxed{A_P=\{10,20,30,40,50,60,70,80, 90,1236\}} .

So, the answer is 10 + 20 + 30 + 40 + 50 + 60 + 70 + 80 + 90 + 1236 = 1686 10+20+30+40+50+60+70+80+90+1236=\boxed{1686} .

Where did you come about knowing about the special property that only the probably considered dull numbers 10, 20, 30, 40, 50, 60, 70, 80, 90, and that the unique number 1236 satisfies the property too, and also, where did you learn about this rule?

Razzi Masroor - 4 years, 5 months ago

Log in to reply

It's was a sudden observation when I noticed this special property of 1236 1236 .

And this 'rule'? If you mean 'the techinques in the proof' by 'rules', then there wasn't a single source. I did the proof myself; but I learned 'induction', 'meaning and significance of necessary-sufficient conditions' from here and there.

Muhammad Rasel Parvej - 4 years, 5 months ago

nice way of introducing the problem

Ashwath Bhat - 4 years, 5 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...