Conservation of Linear momentum gives\begin{equation} m_{1}v_{1}=m_{1}v_{1}^{\prime }+m_{2}v_{2}^{\prime }\tag{1} \end{equation}
We now move the m_{1} terms to one side,\begin{align} v_{1}-v_{1}^{\prime } & =\gamma v_{2}^{\prime }\tag{C1}\\ v_{1}^{2}-\left ( v_{1}^{\prime }\right ) ^{2} & =\gamma \left ( v_{2}^{\prime }\right ) ^{2}\tag{C2} \end{align}
Dividing (2) by (1), using long division (this step is tricky, must be careful), gives\begin{align} \frac{v_{1}^{2}-\left ( v_{1}^{\prime }\right ) ^{2}}{v_{1}-v_{1}^{\prime }} & =v_{2}^{\prime }\nonumber \\ v_{1}+v_{1}^{\prime } & =v_{2}^{\prime }\tag{3} \end{align}
We now replace v_{2}^{\prime } in (C1) with what (3) giving\begin{align} v_{1}-v_{1}^{\prime } & =\gamma \left ( v_{1}+v_{1}^{\prime }\right ) \nonumber \\ v_{1}-\gamma v_{1} & =\gamma v_{1}^{\prime }+v_{1}^{\prime }\nonumber \\ v_{1}\left ( 1-\gamma \right ) & =v_{1}^{\prime }\left ( 1+\gamma \right ) \nonumber \\ v_{1}^{\prime } & =v_{1}\frac{\left ( 1-\gamma \right ) }{\left ( 1+\gamma \right ) }\tag{4} \end{align}
We achieved our goal of finding v_{1}^{\prime } in terms of v_{1}. Now to answer the question. The question is asking to find\begin{equation} \Delta =\frac{T_{1}-T_{1}^{\prime }}{T_{1}}\tag{5} \end{equation}
Using (4) into the above gives \Delta =\frac{v_{1}^{2}-\left ( v_{1}\frac{\left ( 1-\gamma \right ) }{\left ( 1+\gamma \right ) }\right ) ^{2}}{v_{1}^{2}}
Simplifying\begin{align*} \Delta & =\frac{4m_{1}m_{2}}{\left ( m_{1}+m_{2}\right ) ^{2}}\\ & =4\frac{m_{1}m_{2}}{\left ( m_{1}+m_{2}\right ) }\frac{1}{\left ( m_{1}+m_{2}\right ) } \end{align*}
But m=\frac{m_{1}m_{2}}{m_{1}+m_{2}} which is the reduced mass, and M=m_{1}+m_{2}. So the above becomes \Delta =\frac{4m}{M}
Part(a)
Let v_{1} be the speed in the lower circular orbit. Let v_{2} be the speed at the perigee just after speed boost. Let GM\equiv \mu . Since\begin{align*} v_{1} & =\sqrt{\frac{\mu }{r_{A}}}\\ v_{2} & =\sqrt{\mu \left ( \frac{2}{r_{A}}-\frac{1}{a}\right ) } \end{align*}
Where a=\frac{r_{A}+r_{B}}{2}, then \frac{v_{2}}{v_{2}} can now be evaluated\begin{align*} \frac{v_{2}}{v_{1}} & =\frac{\sqrt{\mu \left ( \frac{2}{r_{A}}-\frac{1}{a}\right ) }}{\sqrt{\frac{\mu }{r_{A}}}}\\ & =\sqrt{\frac{\mu \left ( \frac{2}{r_{A}}-\frac{1}{\frac{r_{A}+r_{B}}{2}}\right ) }{\frac{\mu }{r_{A}}}}\\ & =\sqrt{r_{A}\left ( \frac{2}{r_{A}}-\frac{2}{r_{A}+r_{B}}\right ) }\\ & =\sqrt{2-\frac{2r_{A}}{r_{A}+r_{B}}}\\ & =\sqrt{\frac{2\left ( r_{A}+r_{B}\right ) -2r_{A}}{r_{A}+r_{B}}}\\ & =\sqrt{\frac{2r_{B}}{r_{A}+r_{B}}} \end{align*}
Part(b)
Using the period for an ellipse given in the formulas and dividing this by half, since we are looking for half the period, then\begin{align*} T_{p} & =\pi \sqrt{\frac{a^{3}}{G\left ( m_{1}+M_{earth}\right ) }}\\ & =\pi \sqrt{\frac{\left ( \frac{r_{A}+r_{B}}{2}\right ) ^{3}}{G\left ( m_{1}+M_{e}\right ) }} \end{align*}
Assuming the mass of the satellite (m_{1}) is much smaller than M_{earth}, then the above becomes T_{p}=\frac{\pi }{\sqrt{GM_{e}}}\sqrt{\left ( \frac{r_{A}+r_{B}}{2}\right ) ^{3}}
Part(c)
The time it takes B to travel one circle \left ( 2\pi \right ) is T_{c}=2\pi \sqrt{\frac{r_{B}^{3}}{GM}}
Therefore \theta _{0}=\pi \left ( 1-\sqrt{\left ( \frac{r_{A}+r_{B}}{2r_{B}}\right ) ^{3}}\right )