1: % ---Conclusion---------------------------------------------------

2: \label{chp:5}

3: 

4: We have generalized the work of Smoller and Temple given

5: in~\cite{st94} and~\cite{st95} in the sense that we can include

6: lightlike shock surfaces. In defining a more general second

7: fundamental form, by replacing the normal vector with a transverse

8: vector satisfying the jump conditions~\eqref{eq:njump}

9: and~\eqref{eq:njump1}, we were able to overcome the breakdown of

10: the standard second fundamental form for lightlike hypersurfaces.

11: Then we used this generalized second fundamental form, in an

12: analogous way to how Smoller and Temple applied the standard

13: second fundamental form in~\cite{st94}, to obtain a theory for the

14: lightlike case. In the process, we introduced a modified Gaussian

15: Skew coordinate system. This theory yielded the unexpected result

16: of having to include the condition that the metric be $C^2$ on the

17: spacelike subspace of $\tans$.

18: 

19: Then we were able to construct the an exact, spherically symmetric

20: shock-wave solution of the Einstein equations which propagates at

21: the speed of light. No quantities are moving at the speed of light

22: except the shock. Although, our solution was consistent with

23: Smoller and Temple's results in~\cite{st95}, there was the

24: unexpected difference in the value of $\sigma_2$. In the

25: non-lightlike computation given~\cite{st95} the value for $\sigma$

26: in the lightlike limit was

27:   $$\sigma_2 \approx 0.745,$$

28: while our lightlike computations yielded

29:   $$\sigma_2 \approx

30:   0.63442.$$

31: From this we conclude that the limit of the Smoller-Temple

32: subluminal solution as it tends to the speed of light is not equal

33: to our solution propagating at the speed of light.

34: 

35: Furthermore, we showed that in this exact solution the pressure

36: and density are finite on each side of the shock throughout the

37: solution, the sound speeds, on each side of the shock, are

38: constant and subluminous. Moreover, the pressure and density are

39: smaller at the leading edge of the shock which is consistent with

40: the entropy conditions in classical gas dynamics~\cite{lax, smol}.

41: 

42: 

43: \section{Summary of Contributions}

44: In the lightlike shock matching theory of Chapter~\ref{chp:3} we

45: applied the notion of a generalized second fundamental form given

46: by Barrab\`{e}s and Israel in~\cite{isr91} into the shock matching

47: framework of Smoller and Temple~\cite{st94,st95}. The generalized

48: second fundamental form $\K$ in~\cite{isr91} was given in a scalar

49: form, and we modified it so that $\K$ is mapping that takes

50: tangent vectors on the surface to tangent vectors on surface. Then

51: we based the analysis on a Gaussian skew type coordinate system

52: which we had to modify so that we could incorporate the

53: generalized second fundamental form into the shock matching

54: theory. Most of the supporting lemmas, and each of the theorems

55: in~\cite{st94} had to be modified and proved again in the context

56: of this new Modified Gaussian Skew (MGS) coordinate system and the

57: generalized second fundamental form. This yielded the unexpected

58: result of adding the extra condition in our main result,

59: Theorem~\ref{thm:main}, that the metric inner product on the

60: spacelike subspace of $\tans$ to be $C^2$ in order for the

61: Rankine-Hugoniot jump conditions to hold. This condition was

62: already satisfied for the spherically case in

63: Theorem~\ref{thm:sphere}.

64: 

65: Then in Chapter~\ref{chp:4} we constructed a new exact shock-wave

66: solution moving at the speed of light. In order to do this we had

67: to construct appropriate transverse vectors $N$, and apply our

68: extension of the subluminal theory to the specific example of the

69: matched FRW/TOV metrics. This included significant modification of

70: the equations expressing the Rankine-Hugoniot jump conditions

71: given in equation~\eqref{eq:strenj}. This yielded the difference

72: in the limit of the Smoller-Temple subluminal solutions as they

73: tended to the speed of light with the actual solution propagating

74: at the speed of light. Lastly, we incorporated the modifications

75: of the subluminal shock solutions to show that our new exact

76: solution was a crossing shock.

77: