Mummy Carbon Dating

An Egyptian mummy is carbon-dated and compared with a scroll found in the Dead Sea, which is known to be 2000 years old. The X 14 X 2 2 14 C : X 12 X 2 2 12 C \ce{^{14}C} \text{ : } \ce{^{12}C} ratio of the mummy is found to be 2 3 \frac{2}{3} that of the scroll. How old is the mummy (rounded to an integer number of years)?

Details and Assumptions:

  • X 14 X 2 2 14 C \ce{^{14} C} is radioactive with a half-life of 6000 years (it decays into nitrogen). X 12 X 2 2 12 C \ce{^{12} C} is a stable isotope, so the amount of X 12 X 2 2 12 C \ce{^{12} C} in a sample remains constant over time.

  • The ratio X 14 X 2 2 14 C : X 12 X 2 2 12 C \ce{^{14}C} \text{ : } \ce{^{12}C} in an organism (plant or animal) at the time of its death is equal to the ratio of the isotopes in the air it breathed in. This ratio can be assumed to be constant over time.


The answer is 5510.

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.

8 solutions

David Vreken
Mar 18, 2018

The half-life formula is A = A 0 ( 1 2 ) t h A = A_0(\frac{1}{2})^{\frac{t}{h}} , where A A is the final amount, A 0 A_0 is the starting amount, t t is the time, and h h is the half-life.

After 2000 2000 years, the scroll will only have ( 1 2 ) 2000 6000 (\frac{1}{2})^{\frac{2000}{6000}} of its original X 14 X 2 2 14 C \ce{^{14}C} remaining.

Since the mummy has 2 3 \frac{2}{3} of that, we have 2 3 ( 1 2 ) 2000 6000 = ( 1 2 ) x 6000 \frac{2}{3}(\frac{1}{2})^{\frac{2000}{6000}} = (\frac{1}{2})^{\frac{x}{6000}} .

Solving for x x gives x = 6000 ln ( 2 3 ( 1 2 ) 1 3 ) ln ( 1 2 ) 5509.8 x = \frac{6000 \ln (\frac{2}{3}(\frac{1}{2})^{\frac{1}{3}})}{\ln(\frac{1}{2})} \approx \boxed{5509.8} .

The % is superfluous.

A Former Brilliant Member - 3 years, 2 months ago

Log in to reply

Thanks, you are correct. I fixed it.

David Vreken - 3 years, 2 months ago

Sir,is there a way to prove the above mentioned formula??

erica phillips - 3 years, 2 months ago

Log in to reply

A half-life means that an object loses half of that element in that period of time. So if an object starts with A 0 A_0 of an element, then after 1 1 half-life there is 1 2 A 0 \frac{1}{2}A_0 of that element remaining, after 2 2 half-lifes there is 1 2 1 2 A 0 \frac{1}{2}\frac{1}{2}A_0 of that element remaining, and so on, so that in general after n n half-lifes there is ( 1 2 ) n A 0 (\frac{1}{2})^nA_0 of that element remaining. The total time t t would have n n half-lifes, so n = t h n = \frac{t}{h} . Substituting this n n value gives A = A 0 ( 1 2 ) t h A = A_0(\frac{1}{2})^{\frac{t}{h}} .

David Vreken - 3 years, 2 months ago

Log in to reply

Thank u so much!!!

erica phillips - 3 years, 2 months ago

Had, the same approach. ‘This ratio can be assumed to be constant over time.’ is a useless, misleading piece of information.

R Mathe - 3 years, 2 months ago

I might be missing something, but I think that the C14:C12 ratio being 2/3 that of the scroll is different from the percentage of C14 being 2/3 that of the scroll. In the latter case, you're comparing the quantity of C12 to the total quantity of carbon.

Samuel Li - 3 years, 2 months ago

@erica phillips There is no half life in nature: that would involve atoms co-ordinating with each other and saying: okay you 50% must now decay. Doesn’t happen. The standard model views each atom as an independent automata with two states—non-yet-decayed (1) or decayed (0)—and behaving completely independently, but governed by identical rules. The rule for each atom is also simple and clever: each atom decays (changes state from 1 to 0) according to a stochastic process governed by a memoryless distribution, namely the exponential distribution .

Let’s mathematise this. Let X i ( t ) X_{i}(t) be the { 0 , 1 } \{0,1\} -valued state saying whether or not Atom i i at time t t has decayed. By model assumption, the X i ( t ) X_{i}(t) are iid rv (independent identically distributed random variables). That is X i ( t ) X ( t ) X_{i}(t)\sim X(t) for all i i and the rv's are pairwise independent. The distribution is given by P [ X ( t ) = 1 ] = e λ t \mathbb{P}[X(t) = 1] = e^{-\lambda t} and P [ X ( t ) = 0 ] = 1 e λ t \mathbb{P}[X(t) = 0] = 1-e^{-\lambda t} . This arises from the exponential distribution (for the corresponding decay-time rv), where λ \lambda is a constant dependent on the type of atom. It holds that 1 λ = \frac{1}{\lambda}= expected time of decay. In particular E [ X ( t ) ] = 1 e λ t + 0 ( 1 e λ t ) = e λ t \mathbb{E}[X(t)] = 1\cdot e^{-\lambda t} + 0\cdot (1-e^{-\lambda t}) = e^{-\lambda t} is the expected value.

Under this setup, the criteria for the strong law of large numbers are satisfied (look up what this is). The strong law says

1 N i = 1 N X i ( t ) E [ X ( t ) ] = e λ t \frac{1}{N}\sum_{i=1}^{N}X_{i}(t)\longrightarrow \mathbb{E}[X(t)] = e^{-\lambda t}

almost surely for N N\longrightarrow\infty . Thus for N : = A 0 N:=A_{0} a large total number of atoms being viewed, it holds that 1 N i = 1 N X i ( t ) e λ t \frac{1}{N}\sum_{i=1}^{N}X_{i}(t)\approx e^{-\lambda t} . Note that

i = 1 N X i ( t ) = { i 1 i N , X i ( t ) = 1 } = A ( t ) \sum_{i=1}^{N}X_{i}(t)=|\{i\mid 1\leq i\leq N,~X_{i}(t)=1\}|=A(t)

is the number of atoms not-yet decayed at time t t . Hence A ( t ) / A 0 e λ t A(t)/A_{0}\approx e^{-\lambda t} . Hence A ( t ) A 0 e λ t A(t)\approx A_{0}e^{-\lambda t} . Setting h : = log ( 2 ) λ h:=\frac{\log(2)}{\lambda} , one has A ( t ) A 0 ( 1 2 ) t / h A(t)\approx A_{0}(\frac{1}{2})^{t/h} as simplification. The \approx means, that this equation holds with arbitrary levels of accuracy for increasing A 0 A_{0} -values.

And it really is so, that equality almost surely never holds: A ( t ) A(t) does not have an deterministic value at any time (except t = 0 t=0 ). It is a rv with distribution A ( t ) Bin ( A 0 , e λ t ) A(t)\sim\text{Bin}(A_{0},e^{-\lambda t}) . It has variance Var ( A ( t ) ) = A 0 e λ t ( 1 e λ t ) \text{Var}(A(t))=A_{0}e^{-\lambda t}(1-e^{-\lambda t}) and expectation

E [ A ( t ) ] = A 0 e λ t = A 0 ( 1 2 ) t / h . \mathbb{E}[A(t)]=A_{0}e^{-\lambda t}=A_{0}(\frac{1}{2})^{t/h}.

If one scales, then Var ( A ( t ) / A 0 ) = e λ t ( 1 e λ t ) / A 0 0 \text{Var}(A(t)/A_{0})=e^{-\lambda t}(1-e^{-\lambda t})/A_{0}\longrightarrow 0 for A 0 A_{0}\longrightarrow\infty , so that the distribution of A ( t ) / A 0 A(t)/A_{0} gets evermore concentrated around the expected value E [ A ( t ) / A 0 ] = e λ t = ( 1 2 ) t / h \mathbb{E}[A(t)/A_{0}]=e^{-\lambda t}=(\frac{1}{2})^{t/h} . Since in many cases working with the scaled value (the %) is more useful, this justifies replacing the non-deterministic value A ( t ) A(t) by its expected value, which is a deterministic function of t t .

Fazit: There is no halving-process in reality itself. It’s a stochastic process , which however can be approximated by a halving process, and can only be exactly described by a halving-process if one replaces A ( t ) A(t) by the expected value E [ A ( t ) ] \mathbb{E}[A(t)] , which, as discussed in the last paragraph, is under natural circumstances justified.

R Mathe - 3 years, 2 months ago

Log in to reply

Thank u so much for taking the trouble to explain the idea vividly!!!Thanks again:)

erica phillips - 3 years, 2 months ago

Just very old

Hans Zuidervaart - 3 years, 1 month ago
Joseph Giri
Mar 18, 2018

The half-life formula is given by A = A 0 ( 1 2 ) t h A = A_0 (\frac{1}{2}) ^\frac{t}{h} , where A A is the final amount, A 0 A_0 is the initial amount, t t is the time that has passed, and h h is the half-life of the item being measured.

Using subscripts to denote different samples,

A 1 = A 0 ( 1 2 ) t 1 h A_1 = A_0 (\frac{1}{2}) ^\frac{t_1}{h} and A 2 = A 0 ( 1 2 ) t 2 h A_2 = A_0 (\frac{1}{2}) ^\frac{t_2}{h}

Let k k be the proportion of X 12 X 2 2 12 C \ce{^12C} between the samples. Hence, A 1 = k A 2 A_1 = kA_2

Equating and simplifying, we end with the relationship t 1 = t 2 h ln k ln 2 t_1=t_2-\frac{h\ln k}{\ln 2} . We can now compare any 2 samples.

Plugging in the values t 2 = 2000 t_2=2000 , h = 6000 h=6000 and k = 2 3 k= \frac{2}{3} ,

t 1 = 2000 6000 ln 2 3 ln 2 5509.8 t_1=2000-\frac{6000 \ln \frac{2}{3}}{\ln 2}\approx \boxed{5509.8}

Erik Shaw
Mar 19, 2018

All the answers are good. I tend to use exponential functions to work with decay, which is basically just a notation preference. You can model exponential decay with the formula N ( t ) = N 0 e t / τ N(t) = N_0 e^{-t/\tau} where N ( t ) N(t) is the number of 14 C ^{14} C atoms at time t t , N 0 N_0 is the number at t = 0 t=0 , and τ \tau is the exponential time constant. We are given that N 0 2 = N 0 e 6000 / τ l n ( 1 2 ) = 6000 τ τ = 6000 l n ( 1 2 ) \frac{N_0}{2} = N_0 e^{-6000/\tau}\ \Rightarrow ln(\frac{1}{2}) = -\frac{6000}{\tau}\ \Rightarrow \tau = -\frac{6000}{ln(\frac{1}{2})} . We are then given that 2 3 e 2000 / τ = e x / τ x τ = l n ( 2 3 e 2000 / τ ) = 2000 τ + l n ( 2 3 ) x = 2000 l n ( 2 3 ) τ = 2000 + 6000 l n ( 2 3 ) l n ( 1 2 ) 5510 \frac{2}{3} e^{-2000/\tau} = e^{-x/\tau}\ \Rightarrow -\frac{x}{\tau} = ln(\frac{2}{3} e^{-2000/\tau}) = -\frac{2000}{\tau} + ln(\frac{2}{3})\ \Rightarrow x = 2000 - ln(\frac{2}{3})\tau = 2000 + 6000\frac{ln(\frac{2}{3})}{ln(\frac{1}{2})} \approx 5510 .

Andrew Kerr
Mar 23, 2018

log 1 2 \frac{1}{2} ( 2 3 \frac{2}{3} ) * 6000 + 2000 = 5510 years

I've been out of school for a while so here's how I did it:

1) Let A be the exponential decay constant, such that after one half-life exactly one-half of the 14C remains.

2) Let B be the number of years necessary for exactly two-thirds of the original amount of 14C to remain.

Therefore from the problem we can state:

A 6000 = 1 2 A^{6000} = \frac{1}{2} , and

A B = 2 3 A^{B}= \frac{2}{3}

We can take the log of both sides to get

6000 l o g A = l o g 1 2 6000log{A}=log{\frac{1}{2}} , and

B l o g A = l o g 2 3 Blog{A}=log{\frac{2}{3}}

We solve the system for B to get:

B = 6000 l o g 2 3 l o g 1 2 3510 B=\frac{6000\cdot log{\frac{2}{3}}}{log{\frac{1}{2}}}\approx3510 years.

This is how much older the mummy is than the scroll. The scroll is already known to be 2000 years old and so this makes the mummy 5510 years old.

Tadeu Manoel
Mar 21, 2018

Just solve 2^(-x/6000)=2/3*2^(-y/6000), for y=2000

Klara Balic
Mar 25, 2018

Let us define S C S_{\ce{C}} and M C M_{\ce{C}} as the scroll's and mummy's respective X 14 X 2 2 14 C : X 12 X 2 2 12 C \ce{^{14}C}:\ce{^{12}C} ratios.

The function for the radioactive decay of X 14 X 2 2 14 C \ce{^{14}C} , which is at the same time the function for the ratio of X 14 X 2 2 14 C : X 12 X 2 2 12 C \ce{^{14}C}:\ce{^{12}C} , is N ( t ) = N 0 x t N(t) = N_0 * x^t , where t t is the time given in years. The starting value N 0 N_0 is unimportant as the problem concerns ratios, and x x is unknown.

We know that x 6000 = 1 2 x^{6000}=\frac{1}{2} and can from there reason that:

x = 1 2 6000 x = \sqrt[6000]{\frac{1}{2}}

x = 1 2 6000 x =\frac {1} {\sqrt[6000]{2}}

x = 2 1 6000 x = {2}^ {-\frac{1}{6000}}

This makes the formula for radioactive decay N ( t ) = 2 t 6000 N(t) = {2}^ {-\frac{t}{6000}}

S C = N ( 2000 ) = 2 2000 6000 = 1 2 3 S_{\ce{C}} = N({2000}) = {2}^ {-\frac{2000}{6000}} = \frac {1}{\sqrt[3]{2}}

For the age a a of the mummy, we can take the equation:

M C = 2 a 6000 = 2 3 1 2 3 M_{\ce{C}}= {2}^{-\frac{a}{6000}} = \frac23 * \frac{1}{\sqrt[3]{2}}

a ln ( 2 1 6000 ) = ln ( 2 3 2 3 ) a * \ln({2}^{-\frac{1}{6000}}) = \ln(\frac {2}{{3}\sqrt[3]{2}})

a = ln ( 2 3 2 3 ) ln ( 2 1 6000 ) = 5509.775 5510 a = \frac{\ln(\frac {2}{{3}\sqrt[3]{2}})}{\ln({2}^{-\frac{1}{6000}})} = 5509.775\ldots \approx \boxed {5510}

Pieter van Nes
Mar 21, 2018

log(1/2)/log(2/3)=3510

3510+2000=5510 years.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...