The power of two is more fun than the power of one

2 1 2 2 + 2 3 2 4 + 2 5 2 6 + 2 7 + 2 2013 2^{1}-2^{2}+2^{3}-2^{4}+2^{5}-2^{6}+2^{7} -\ldots +2^{2013}

Let a a be the value of the expression above. Find the last two digits of a . a.


The answer is 62.

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.

11 solutions

The sum can be written as

a = 2 ( 2 0 2 1 + 2 2 2 3 + . . . . . + 2 2012 ) = a = 2*(2^{0} - 2^{1} + 2^{2} - 2^{3} + ..... + 2^{2012}) =

2 ( 2 ) 2013 1 ( 2 ) 1 = 2 3 ( 2 2013 + 1 ) 2*\dfrac{(-2)^{2013} - 1}{(-2) - 1} = \dfrac{2}{3} (2^{2013} + 1) .

Now in order to find the last two digits of a a we will need to determine a m o d 100 a\mod 100 . Now 2 22 2 2 m o d 100 2^{22} \equiv 2^{2}\mod 100 , i.e., the remainders m o d 100 \mod 100 have a 'period' of 20 20 , so

2 2002 2 2 m o d 100 2 2013 2 13 m o d 100 92 m o d 100 2^{2002} \equiv 2^{2}\mod 100 \Longrightarrow 2^{2013} \equiv 2^{13}\mod 100 \equiv 92\mod 100 .

Therefore 2 3 ( 2 2013 + 1 ) 2 3 ( 93 ) 62 m o d 100 \dfrac{2}{3}(2^{2013} + 1) \equiv \dfrac{2}{3}(93) \equiv 62\mod 100 .

Thus the last two digits of a a are 62 \boxed{62} .

Although your final answer is right, it's impractical to actually calculate the "period" of 20 20 as you can easily make plenty of arithmetic errors. And your "period" is not a rigorous proof, you have only observed that pattern repeats after 20 terms but you didn't explain why it happens.

What I would prefer to do is set b = 3 a 2 = 2 2014 b = 3a -2 = 2^{2014} then apply Chinese Remainder Theorem for modulo 100 = 4 × 25 100 = 4 \times 25 . Noting that 4 , 25 4,25 are coprime. We have the linear congruences: b 0 ( m o d 4 ) b \equiv 0 \pmod 4 , and for modulo 25 25 : b 2 2014 m o d ϕ ( 25 ) 2 14 ( 2 7 ) 2 3 2 9 b \equiv 2^{2014 \bmod{\ \phi(25)}} \equiv 2^{14} \equiv (2^7)^2 \equiv 3^2 \equiv 9 .

By CRT, we get: b 84 ( m o d 100 ) 3 a 86 ( m o d 100 ) b\equiv 84 \pmod {100} \Rightarrow 3a \equiv 86 \pmod{100}

3 a 86 + 100 ( m o d 100 ) a 62 ( m o d 100 ) \Rightarrow 3a \equiv 86 + 100 \pmod{100} \Rightarrow a \equiv \boxed{62} \pmod {100} .

Pi Han Goh - 6 years, 2 months ago

Log in to reply

Can you explain the mod 25 part a bit more elaborately. I couldn't understand how did Euler's totient function phi has come in here?

Puneet Pinku - 5 years ago

Log in to reply

Log in to reply

@Pi Han Goh I know them, I just couldn't understand the application part

Puneet Pinku - 5 years ago

Log in to reply

@Puneet Pinku Write down the exact step where you don't understand.

Pi Han Goh - 5 years ago

Log in to reply

@Pi Han Goh Calculation of mod25 part?

Puneet Pinku - 5 years ago

Log in to reply

@Puneet Pinku In the wiki I've provided above, a ϕ ( n ) 1 ( m o d n ) a^{\phi(n) } \equiv 1 \pmod n , where gcd ( a , n ) = 1 \gcd(a,n) = 1 . In this case, a = 2 , n = 25 a = 2, n = 25 .

Pi Han Goh - 5 years ago

Log in to reply

@Pi Han Goh So we get 2 20 m o d 25 2^{20} \equiv mod 25 . After that...

Puneet Pinku - 5 years ago

Log in to reply

@Puneet Pinku NO. it's 2 20 1 ( m o d 25 ) 2^{20} \equiv 1 \pmod{25} . Now, we want to find 2 2014 m o d 25 2^{2014} \bmod{25} , which is just 2 2014 m o d ( ϕ ( 25 ) ) m o d 25 2^{2014 \bmod{(\phi(25))}} \bmod {25} by Euler's Theorem.

Pi Han Goh - 5 years ago

Log in to reply

@Pi Han Goh Ah! I see.. Now how did the 2^14 turn into 3^2

Puneet Pinku - 5 years ago

Log in to reply

@Puneet Pinku Follow the steps that I've given above.

Pi Han Goh - 5 years ago

Log in to reply

@Pi Han Goh Okay! But why is 100 added to the remainder after getting 3a m o d \equiv mod 25 is obtained.

Puneet Pinku - 5 years ago

Log in to reply

@Puneet Pinku If A B ( m o d C ) A \equiv B \pmod C , then A B + C ( m o d C ) A \equiv B + C \pmod C is true as well.

Pi Han Goh - 5 years ago

Log in to reply

@Pi Han Goh Thankyou very much I now understand it completely.

Puneet Pinku - 5 years ago

Euler's theorem can be used if a a and n n are co-prime.

Kartik Sharma - 6 years, 5 months ago

Same query as Kartik According to the Theorem a ϕ ( n ) 1 m o d n a^{\phi (n)}\equiv 1\mod n

Both a and n should be coprime. However 2 and 100 are not. How does the solution work still work.

It should be noted that 2 40 76 m o d 100 2^{40} \equiv 76 \mod 100

Great Soution However!

Sualeh Asif - 6 years, 5 months ago

Log in to reply

Yes, you and @Kartik Sharma are correct; that was a silly oversight on my part. It was just lucky that the wrong method produced the right answer. I've found a correct method and have made the appropriate edits. Thanks for pointing out my mistake. :)

Brian Charlesworth - 6 years, 5 months ago

You could use the chinese remainder though to divide the task of finding mod 100 into finding mod 4 and mod 25 respectively, that way 25 and 2 are coprime

Nicolas Nauli - 1 year, 2 months ago
Arjen Vreugdenhil
Oct 25, 2015

I started by noticing ( 2 3 2 2 ) + ( 2 5 2 4 ) = 20 , (2^3-2^2) + (2^5-2^4) = 20, and the next set of four powers of two would be ( 2 7 2 6 ) + ( 2 9 2 8 ) = 2 4 ( 2 3 2 2 + 2 5 2 4 ) = 16 20 = 320 , (2^7-2^6) + (2^9-2^8) = 2^4\cdot(2^3-2^2+2^5-2^4) = 16\cdot 20 = 320, and continuing this pattern we keep having as last digits ...20.

Since there are 2012 of such terms to account for (we skipped 2 1 2^1 ), we would add 2012 / 4 = 503 2012/4 = 503 of these 20s. It is easy to see that this gives last digits ...60.

Finally, adding in 2 1 = 2 2^1 = 2 , we get 62 \boxed{62} .

nice solution

mukesh malav - 5 years, 2 months ago

that was an awsomest answer, the simplest! cheers!

Vivek Panade - 4 years, 6 months ago

Math is fun.Let it be enjoyed.

shital jha - 4 years ago

Wow, clear for who doesn't learn any theorem!

Kelvin Hong - 3 years, 9 months ago

2 1 2 2 + 2 3 + . . . + 2 2013 = 2 1 + 2 2 1 + 2 2 2 + . . + 2 2 1006 2^1-2^2+2^3+...+2^{2013}=2^1+2^{2*1}+2^{2*2}+..+2^{2*1006} It is easy to realize that: 2 2 ( 10 i + 1 ) + 2 2 ( 10 i + 2 ) + 2 2 ( 10 i + 3 ) + . . . + 2 2 ( 10 i + 10 ) m o d 100 = 0 {2^{2*(10*i+1)}+2^{2*(10*i+2)}+2^{2*(10*i+3)}+...+2^{2*(10*i+10)}} \mod 100 = 0 with i=0..99 So: 2 1 2 2 + 2 3 + . . . + 2 2013 m o d 100 2^1-2^2+2^3+...+2^{2013} \mod 100 = 2 1 + 2 2002 + 2 2004 + . . . + 2 2012 m o d 100 = 62 =2^1+2^{2002}+2^{2004}+...+2^{2012} \mod 100=\boxed{62}

Sridhar Sri
Mar 4, 2016

Rohit Sachdeva
Jan 3, 2016

Lets try to sum in the usual way:

S = ( 2 1 + 2 3 + . . . . 2 2013 ) ( 2 2 + 2 4 + . . . . . 2 2012 ) S=(2^{1}+2^{3}+....2^{2013})-(2^{2}+2^{4}+.....2^{2012})

= 2 ( 2 0 + 2 2 + . . . . 2 2012 ) ( 2 0 + 2 2 + . . . . . 2 2012 ) + 1 =2(2^{0}+2^{2}+....2^{2012})-(2^{0}+2^{2}+.....2^{2012})+1

= ( 2 0 + 2 2 + 2 4 + . . . . . 2 2012 ) + 1 =(2^{0}+2^{2}+2^{4}+.....2^{2012})+1

The expression in () is an AP with a=1, r=2², n=1007

Which gives S = 4 1007 1 3 + 1 S=\frac{4^{1007}-1}{3}+1

Observe the following:

4 1 1 3 = 01 \frac{4^{1}-1}{3}=01

4 3 1 3 = 21 \frac{4^{3}-1}{3}=21

4 5 1 3 = 341 \frac{4^{5}-1}{3}=341

4 7 1 3 = . . . 61 \frac{4^{7}-1}{3}=...61

4 9 1 3 = . . . . 81 \frac{4^{9}-1}{3}=....81

4 11 1 3 = . . . . . 01 \frac{4^{11}-1}{3}=.....01

So, 4 1007 1 3 = . . . . . . . 61 \frac{4^{1007}-1}{3}=.......61

Hence S will end in 61+1 = 62 \boxed{62}

Matthew Cox
Jan 8, 2015

Since we only need the last 2 digits, we can do all of our computation in mod 100.

First, we find the residues (mod 100) of the powers of 2, just doubling each time and subtracting 100 if necessary:

2 1 2 ( m o d 100 ) 2^1\equiv 2\pmod{100}

2 2 4 2^2\equiv 4

2 3 8 2^3\equiv 8

2 4 16 2^4\equiv 16

2 5 32 2^5\equiv 32

2 6 64 2^6\equiv 64

2 7 28 2^7\equiv 28

2 8 56 2^8\equiv 56

2 9 12 2^9\equiv 12

2 10 24 2^{10}\equiv 24

2 11 48 2^{11}\equiv 48

2 12 96 2^{12}\equiv 96

2 13 92 2^{13}\equiv 92

2 14 84 2^{14}\equiv 84

2 15 68 2^{15}\equiv 68

2 16 36 2^{16}\equiv 36

2 17 72 2^{17}\equiv 72

2 18 44 2^{18}\equiv 44

2 19 88 2^{19}\equiv 88

2 20 76 2^{20}\equiv 76

2 21 52 2^{21}\equiv 52

2 22 2 2 4 2^{22}\equiv 2^2\equiv 4

We find that every 20 powers of 2, the residue in mod 100 repeats.

Next, we rearrange the original equation to 2 2013 2 1 2 3 2 5 . . . 2 2011 2^{2013}-2^1-2^3-2^5...-2^{2011} . We can do this because 2 n + 1 2^{n+1} is twice 2 n 2^n , so 2 n 2 n + 1 = 2 n 2^n-2^{n+1}=-2^n .

In mod 100, this is equivalent to 2 13 2 1 100 ( 2 3 ) 100 ( 2 5 ) 100 ( 2 7 ) 100 ( 2 9 ) 100 ( 2 11 ) 100 ( 2 13 ) 2^{13}-2^1-100(2^3)-100(2^5)-100(2^7)-100(2^9)-100(2^{11})-100(2^{13})-

101 ( 2 15 ) 101 ( 2 17 ) 101 ( 2 19 ) 100 ( 2 21 ) 2 13 2 1 2 15 2 17 2 19 101(2^{15})-101(2^{17})-101(2^{19})-100(2^{21})\equiv 2^{13}-2^1-2^{15}-2^{17}-2^{19}\equiv

92 2 68 72 88 62 ( m o d 100 ) 92-2-68-72-88\equiv \fbox{62} \pmod{100} .

Leonhard Euler
Apr 4, 2020

First, rewrite the sum as

2 1 2 2 + 2 3 2 4 + 2 5 + 2 2013 = 2 1 + ( 2 2 + 2 3 ) + ( 2 4 + 2 5 ) + + ( 2 2012 + 2 2013 ) = 2 + 2 2 + 2 4 + 2 6 + + 2 2012 = 2 + 4 1 + 4 2 + 4 3 + + 4 1006 . \begin{aligned} & 2^{1}-2^{2}+2^{3}-2^{4}+2^{5}-\ldots +2^{2013}\\ =~& 2^{1}+(-2^{2}+2^{3})+(-2^{4}+2^{5})+\ldots +(-2^{2012}+2^{2013})\\ =~& 2+2^{2}+2^{4}+2^{6}+\ldots +2^{2012}\\ =~& 2+4^{1}+4^{2}+4^{3}+\ldots +4^{1006}. \end{aligned}

Now, note that 4 1 + 4 2 20 ( m o d 100 ) 4^{1}+4^{2}\equiv 20 \pmod{100} . Multiplying by 4 2 4^{2} gives us 4 3 + 4 4 320 20 ( m o d 100 ) 4^{3}+4^{4}\equiv 320\equiv 20 \pmod{100} . Continuing this process yields 4 2 n 1 + 4 2 n 20 ( m o d 100 ) 4^{2n-1}+4^{2n}\equiv 20 \pmod{100} . Therefore

2 + ( 4 1 + 4 2 ) + ( 4 3 + 4 4 ) + + ( 4 1005 + 4 1006 ) 2 + 20 + + 20 503 ( m o d 100 ) = 10062 62 ( m o d 100 ) . \begin{aligned} &~ 2+(4^{1}+4^{2})+(4^{3}+4^{4})+\ldots +(4^{1005}+4^{1006})\\ \equiv &~ 2 + \underbrace{20+\ldots +20}_{503} \pmod{100}\\ =&~ 10062\\ \equiv &~ \boxed{62} \pmod{100}. \end{aligned}

Andrew Amin
Aug 29, 2019

xt=0; a= [4 8 16 32 64 28 56 12 24 48 96 92 84 68 36 72 44 88 76 52]; for y= 2:2013; if rem (y,20) == 0 c= 20 ; else c = rem (y,20); end x=a(c); xt = xt+x;

end disp (rem(xt+2,100))

David Krüger
Apr 6, 2016

Once again, not the best way to find a solution, as it certainly doesn't prove the existence of the solution:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
format long g, clear, clc
b = 0;

for (i = 1:100)
    a = ((-1)^(i-1))*(2^i);
    b = b + a
end

% 2260
% 00012487486362498625

The comments at the bottom, in order, are (respectively) the pattern of the one's place digit, and the pattern of the ten's place digit. This works until until scientific notation kicks in, so i i to 100 is actually quite overkill.

Since the first pattern has 4 numbers, 2013 (2012 divisible by 4) ends on the first number, which is 2.

The second pattern has 20, and after 2000 it ends on the 13th number, which is 6.

So the sum of the sequence will end with '62'

(2^n+2)/3, where n%4=2..... the second last digit when (2^n+2)/3 is ((n-2/4)%5)*(2)...In our case n=2014.... So ((2014-2/4)%5)(2)=(503%5)(2) =(3)(2)=6...Here the last digit of 2^n is 4..by n%4...So (2^n+2) has 6 as its last digit..:. (2^n+2)/3 has 2 as its last digit..So last 2 digits are 62..

Jakub Šafin
Jan 5, 2015

Let's calculate just remainders mod 4 and 25. Mod 4 is easy: the remainder is 2. Mod 25: subtract 1 = 2 0 1=2^0 from the sum and notice that it can be rewritten as k = 0 1006 4 k \sum_{k=0}^{1006}4^k . When we notice that 4 5 = 1024 1 4^5=1024\equiv -1 , we know that any 10 consecutive summands give a remainder of 0, so the whole sum's remainder is k = 0 6 4 k 1 + 4 + 16 + 14 + 6 1 4 = 11 \sum_{k=0}^6{4^k}\equiv 1+4+16+14+6-1-4=11 ; the original sum's remainder then must have been 12.

Chinese remainder theorem guarantees that there's exactly one number 0 , < 100 \ge 0, < 100 that satisfies the two congruences; that number is easily guessed to be 62 (not 12, definitely not 37 or 87).

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...