A beautiful blue light

In 1-d Newtonian mechanics the energy as a function of momentum for a particle is given by E = 1 2 m v 2 E=\frac{1}{2}mv^2 or, in terms of momentum, E = p 2 / ( 2 m ) E=p^2/(2m) . In Einstein's theory of special relativity there is still energy and momentum, and they are still conserved, but the relationship between the two is different. In empty space, the energy of a particle as a function of its momentum p \vec{p} and mass m m is given by E 2 = c 2 p p + m 2 c 4 E^2=c^2 \vec{p} \cdot \vec{p}+m^2c^4 where c c is the speed of light. The photon, the "particle" of light, has no mass - its energy satisfies E = c p E=c|\vec{p}| in empty space. In a medium like water this can change - photons travel slower in a medium and their energy-momentum relation becomes E = c m p E=c_m |\vec{p}| , where c m c_m is the speed of light in the medium. Our question is the following: A very high energy proton with E p r o t 2 = c 2 p p + m p r o t 2 c 4 E_{prot}^2=c^2 \vec{p} \cdot \vec{p}+m_{prot}^2c^4 enters a tank of water. What is the minimum magnitude of the proton's momentum, i.e. p |\vec{p}| , in kg m/s , such that the proton can emit a photon with non-zero momentum and still conserve energy and momentum?

Details and assumptions

  • The speed of light in empty space is 3 × 1 0 8 m/s 3 \times 10^8~\mbox{m/s} .
  • The mass of the proton is 1.67 × 1 0 27 kg 1.67 \times 10^{-27}~\mbox{kg} .
  • The speed of light in water is 2.3 × 1 0 8 m/s 2.3 \times 10^8~\mbox{m/s} .
  • When photon emission begins to occur the initial proton, the final proton, and the emitted photon all travel in the same direction.


The answer is 5.98E-19.

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.

2 solutions

Josh Silverman Staff
Nov 18, 2013

As the noted TV personality/string theorist Brian Greene says, "energy is the most convertible currency in the universe". At the core, this question is about the economics of photon production.

In order to create a photon of momentum p p in the tank of water we must pay the photon factory with the energy p c m pc_m . I.e. the cost of photon momentum (in the denomination of energy) is given by E γ p γ = c m \displaystyle \frac{\partial E_{\gamma}}{\partial p_{\gamma}} = c_m .

Let us assume that the photon is produced at no loss. Under this assumption, the only source of momentum or energy for the photon are the commensurate losses in the slowing proton, therefore any momentum lost by the proton shows up in the proton and, likewise, any momentum lost by the proton shows up in the photon. This will lead us to the minimum momentum the proton must have to radiate a photon.

For every unit of momentum lost by the particle, the loss in energy is given by

E p p p = p p 2 c 2 + m 0 2 c 4 = p c 2 E \displaystyle \frac{\partial E_{\mbox{p}}}{\partial p_{\mbox{p}}} = -\frac{\partial}{\partial p} \sqrt{p^2c^2+m_0^2c^4} = -\frac{pc^2}{E}

In other words, the maximum price that slowing proton can pay to the photon factory is p c 2 E \displaystyle \frac{pc^2}{E} units of energy per unit of momentum.

Let's play with this expression a bit. The energy of the proton E E can also be written as E = m 0 γ c 2 E = m_0\gamma c^2 where γ \gamma is the Lorentz factor 1 v 2 / c 2 \sqrt{1-v^2/c^2} (not to be confused with the subscript I used to denote the photon above). Another way to write the proton momentum is given by p = m 0 v γ p = m_0 v \gamma .

Now we can see that the price that the slowing proton can afford to pay for photons is given by

E p p p = m 0 v γ c 2 m 0 γ c 2 = v \displaystyle \frac{\partial E_{\mbox{p}}}{\partial p_{\mbox{p}}} = \frac{m_0v\gamma c^2}{m_0\gamma c^2} = v

Putting it altogether, the fixed price of photon momentum in the denomination of energy is given by c m c_m . The maximum price that the proton can afford to pay is given by its velocity, v v . A fast proton can afford to pay more than a slow proton. Obviously, the proton can't make any photons if it can't afford c m c_m so at the very least it must be able to pay c m c_m , i.e. it must be traveling with the velocity c m c_m , at the very least, with respect to the water tank. This shows that v c m v \geq c_m .

The minimum momentum required to radiate the blue light is then p ( c m ) = m 0 c m γ ( c m ) = m 0 c m 1 c m 2 / c 2 6 × 1 0 19 kg m/s \displaystyle p(c_m) = m_0c_m\gamma(c_m) = \frac{m_0 c_m}{\sqrt{1-c_m^2/c^2}} \approx \boxed{ 6\times 10^{-19}\mbox{ kg m/s}} .

*Note that this cutesy (maybe too cute) mapping of photon production onto the buying of photon momentum at the market for energy dollars is really just the combined principles of the conservation of momentum and energy. Matching the loss of energy relative to momentum loss in the proton to the gain in energy relative to momentum gain in the photon ensures that both energy and momentum are conserved in the reaction.

Mark Hennings
Nov 20, 2013

All the motion is in the same straight line, so this becomes a scalar, not vector, problem. Suppose that the momentum of the proton is p p initially and q q finally, and that the momentum of the photon is r r , where p , q , r > 0 p,q,r > 0 . Then conservation of energy and momentum tell us that c 2 p 2 + m 2 c 4 = c 2 q 2 + m 2 c 4 + c m r p = q + r \begin{array}{rcl} \sqrt{c^2p^2 + m^2c^4} & = & \sqrt{c^2q^2 + m^2c^4} + c_mr \\ p & = & q + r \end{array} If we write γ = c m c \gamma = \frac{c_m}{c} , then p 2 + m 2 c 2 = q 2 + m 2 c 2 + γ ( p q ) γ 2 ( p q ) 2 = [ p 2 + m 2 c 2 q 2 + m 2 c 2 ] 2 = p 2 + q 2 + 2 m 2 c 2 2 ( p 2 + m 2 c 2 ) ( q 2 + m 2 c 2 ) \begin{array}{rcl} \sqrt{p^2 + m^2c^2} & = & \sqrt{q^2 + m^2c^2} + \gamma(p-q) \\ \gamma^2(p-q)^2 & = & \big[\sqrt{p^2+m^2c^2} - \sqrt{q^2+m^2c^2}\big]^2 \\ & = & p^2 + q^2 + 2m^2c^2 - 2\sqrt{(p^2+m^2c^2)(q^2+m^2c^2)} \end{array} and so 4 ( p 2 + m 2 c 2 ) ( q 2 + m 2 c 2 ) = [ p 2 + q 2 γ 2 ( p q ) 2 + 2 m 2 c 2 ] 2 4(p^2+m^2c^2)(q^2+m^2c^2) \; = \; \big[p^2 + q^2 - \gamma^2(p-q)^2 + 2m^2c^2\big]^2 which simplifies to ( p q ) 2 { ( 1 γ 2 ) [ ( p + q ) 2 γ 2 ( p q ) 2 ] 4 γ 2 m 2 c 2 } = 0 (p-q)^2\left\{ (1-\gamma^2)\big[(p+q)^2 - \gamma^2(p-q)^2\big] - 4\gamma^2m^2c^2\right\} \; = \; 0 We are interested in solutions where p q = r > 0 p-q=r >0 , and so we need to solve ( 1 γ 2 ) [ ( p + q ) 2 γ 2 ( p q ) 2 ] = 4 γ 2 m 2 c 2 (1-\gamma^2)\big[(p+q)^2 - \gamma^2(p-q)^2\big] \; = \; 4\gamma^2m^2c^2 This is the equation of a rotated hyperbola in the ( p , q ) (p,q) plane, and we are interested in solutions for which p > q > 0 p > q > 0 . Thus we require p > p m i n p > p_\mathrm{min} , where p m i n > 0 p_\mathrm{min} > 0 and ( 1 γ 2 ) [ ( p m i n + p m i n ) 2 γ 2 ( p m i n p m i n ) 2 ] = 4 γ 2 m 2 c 2 ( 1 γ 2 ) p m i n 2 = γ 2 m 2 c 2 \begin{array}{rcl} (1-\gamma^2)\big[(p_\mathrm{min} + p_\mathrm{min})^2 - \gamma^2(p_\mathrm{min} - p_\mathrm{min})^2\big] & = & 4\gamma^2m^2c^2 \\ (1-\gamma^2)p_\mathrm{min}^2 & = & \gamma^2m^2c^2 \end{array} (the line p = q p=q meets the hyperbola in the first quadrant at the point ( p m i n , p m i n ) (p_\mathrm{min},p_\mathrm{min}) ), and hence p m i n = γ m c 1 γ 2 = m c c m c 2 c m 2 p_\mathrm{min} \; = \; \frac{\gamma m c}{\sqrt{1-\gamma^2}} \; = \; \frac{mcc_m}{\sqrt{c^2 - c_m^2}} This gives us the value p m i n = 5.98 × 1 0 19 p_\mathrm{min} = 5.98 \times 10^{-19} kg m/s.

When I saw the question, I got completely confused because I thought that only proton's momentum and energy are conserved. I kept thinking that after proton emits the photon, how would still it conserve the energy. Isn't the statement of the problem wrong in some sense?

Snehal Shekatkar - 7 years, 6 months ago

Someone ought to point out that we are discussing Cerenkov radiation here, and that it tends to be a pretty hue of blue, when visible!

Peter Baumgart - 3 years, 2 months ago

Nice solution (didn't manage to solve this problem myself though)!

Just would like to point out that alternatively, without recognizing that ( 1 γ 2 ) [ ( p + q ) 2 γ 2 ( p q ) 2 ] = 4 γ 2 m 2 c 2 (1-\gamma ^2 ) [(p+q)^2 - \gamma ^2 (p-q)^2 ] = 4 \gamma ^2 m^2 c^2 is a hyperbola, by considering p>q>0, we may also simplify directly that

4 γ 2 m 2 c 2 4 \gamma ^2 m^2 c^2

= ( 1 γ 2 ) [ ( p + q ) 2 γ 2 ( p q ) 2 ] = (1-\gamma ^2 ) [(p+q)^2 - \gamma ^2 (p-q)^2 ]

< ( 1 γ 2 ) [ ( p + q ) 2 ] < (1-\gamma ^2 ) [(p+q)^2]

< ( 1 γ 2 ) [ ( 2 p ) 2 ] < (1-\gamma ^2 ) [(2p)^2] , i.e.

p > γ m c ( 1 γ 2 ) p > \frac{ \gamma m c }{ \sqrt{(1-\gamma ^2 )}} , which is the same result. :D

Jau Tung Chan - 7 years, 6 months ago

Log in to reply

That shows that what I called p m i n p_\mathrm{min} is a lower bound; maybe a little more is needed to confirm that it is a greatest lower bound.

Mark Hennings - 7 years, 6 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...