1:
2: \documentstyle[prl,aps,epsfig,multicol]{revtex}
3: \begin{document}
4: \def\be{\begin{equation}}
5: \def\ee{\end{equation}}
6: \def\bearr{\begin{eqnarray}}
7: \def\eearr{\end{eqnarray}}
8: \def\tc{$T_c~$}
9: \def\tcl{$T_c^{1*}~$}
10: \def\c2{ CuO$_2~$}
11: \def\ruo{ RuO$_2~$}
12: \def\lsco{LSCO~}
13: \def\bi{bI-2201~}
14: \def\tl{Tl-2201~}
15: \def\hg{Hg-1201~}
16: \def\sro{$Sr_2 Ru O_4$~}
17: \def\rc{$RuSr_2Gd Cu_2 O_8$~}
18: \def\mgb{$MgB_2$~}
19: \def\pz{$p_z$~}
20: \def\ppi{$p\pi$~}
21: \def\sqo{$S(q,\omega)$~}
22: \def\tperp{$t_{\perp}$~}
23:
24:
25: \title{Reply to a comment by Peres et al.}
26:
27: \author{ G. Baskaran$^1$ and S.A. Jafari$^{1,2}$\\
28: ${}^1$Institute of Mathematical Sciences,
29: C.I.T. Campus, Chennai 600 113, India \\
30: ${}^2$Department of Physics, Sharif University of Technology,
31: Tehran 11365-9161, Iran}
32:
33: \maketitle
34: \begin{abstract}
35:
36: \end{abstract}
37:
38: \begin{multicols}{2}[]
39:
40: Peres et al. comment\cite{peres} that in our recent work on
41: graphite\cite{gbak1} we get spin-1 collective (spin-1 zero sound)
42: mode only when we use, in addition to `random phase approximation' (RPA),
43: another approximation.
44:
45: To justify our method and also clarify the issue involved, {\bf
46: particularly at low energies}, we discuss our
47: spin-1 zero sound in a fermi liquid theory framework\cite{gbak2} and
48: emphasize an important aspect of Landau's zero sound collective mode
49: analysis. One calculates
50: the two particle forward scattering amplitude defined by Landau
51: $\Gamma(p_1,p_2;k)$ by focusing on the singularity in the zero sound
52: channel. Most importantly, as emphasized by Landau\cite{landau} one considers
53: intermediate states in which the number of `quasi-particles' are not changed.
54: The RPA susceptibility we have computed in our approximation precisely
55: corresponds to a bubble sum of such a process, where the quasi-particle
56: number is conserved in the intermediate states.
57:
58: To elaborate this point we write the `bare' Hubbard Hamiltonian for the
59: ideal graphene sheet in terms of the conduction band electron ($e'$s)
60: and valence band hole ($h$'s) operators with {\bf positive energies}:
61: \bearr
62: H & = & \sum_{k \sigma} \epsilon_k ( e^{\dagger}_{k\sigma} e^{}_{k\sigma} +
63: h^{\dagger}_{k\sigma} h^{}_{k\sigma}) \nonumber \\
64: &+& \frac{U}{N} \sum_{k k'}
65: \gamma^{(1)}_{k,k',q} (e^{\dagger}_{k\uparrow} e^{}_{k+q \uparrow}
66: h^{\dagger}_{k'\downarrow} h^{}_{k'-q\downarrow} + ..)\nonumber \\
67: &+& \frac{U}{N} \sum_{k k'} ( \gamma^{(2)}_{k,k',q}e^{\dagger}_{-k\uparrow}
68: h^{\dagger}_{k+q\uparrow} e^{\dagger}_{-k'\downarrow}
69: h^{\dagger}_{k'-q\downarrow} \nonumber \\
70: &+&
71: \gamma^{(3)}_{k,k',q}e^{\dagger}_{k\uparrow}e^{}_{k-q\uparrow}
72: e^{\dagger}_{-k'\downarrow} h^{\dagger}_{k'-q \downarrow + ..}
73: \eearr
74: Here $ \epsilon_k \equiv t \sqrt{ 1 + 4 \cos \frac{{\sqrt 3}k_x}{2}
75: \cos \frac{k_y}{2} + 4 \cos^2 \frac{k_y}{2} } $ and
76: $\gamma$'s are complex form factors, $|\gamma| \sim 1$.
77: Hubbard interaction terms describe various two body processes:
78: electron-electron, hole-hole, electron-hole scattering, electron
79: decay, vacuum polarization etc.
80: The second line of eqn.(1), $e^{\dagger}e^{}h^{\dagger}h^{}$
81: is the one important for us - it represents scattering between a real
82: electron and a real hole. The term $ e^{\dagger}e^{\dagger}
83: h^{\dagger}h^{\dagger}$ of the third line corresponds to
84: spontaneous creation of two electron-hole pairs out of the vacuum and
85: the second term $e^{\dagger}e^{}h^{\dagger}e^{\dagger}$ represents
86: decay of a real electron into an electron and electron-hole pair.
87: That is, all terms contained in the third line are either processes
88: that renormalize the vacuum or dress the quasi-particles.
89: At the end, if a fermi liquid state is formed, one has `stable' low energy
90: renormalized {\bf quasi particles that neither feel the vacuum polarization
91: processes nor polarize the vacuum}.
92: In a renormalization group sense, the $\gamma^{(2)}$, $\gamma^{(3)}$ etc.
93: terms of eqn.(1) are expected to renormalize to zero, as one approaches a
94: fermi liquid fixed point. For our analysis of spin-1 zero sound in graphite,
95: assumed to be a fermi liquid, we are essentially left with the $\gamma^{(1)}$
96: term with a renormalized ${\tilde U} > 0$. That is, one low energy electron
97: and one hole quasi-particle repeatedly scatter, in a renormalized
98: {\bf frozen fermi liquid vacuum},
99: without intervention of any additional quasi-particles
100: in the intermediate state, in the way described by Landau.
101: This leads to our simple susceptibility expression\cite{gbak1}
102: \be
103: \chi(q,\omega) =
104: \frac{\chi_0 (q,\omega)}{1 - {\tilde U} \chi_0(q,\omega)}
105: \ee
106: that has poles corresponding to our predicted spin-1 collective mode
107: branch, below the particle-hole continuum.
108:
109: \begin{figure}[h]
110: \epsfxsize 6cm
111: \centerline {\epsfbox{bantfig1.eps}}
112: \caption{A diagram that is contained in Peres et al.'s bubble summation.
113: Intermediate state A is a 1-electron and 1-hole state; B is a 3-electron
114: and 3-hole state. At low energies, in a Landau theory such diagrams with
115: more than 1-electron and 1-hole in the intermediate states do not exist,
116: because $U\gamma^{(2)}$ is renormalized to zero/low values.}
117: \end{figure}
118:
119: Peres et al., have, on the other hand, calculated a `2-sub lattice
120: susceptibility tensor', which uses the full unrenormalized Hubbard
121: interaction terms
122: [lines 2,3 and 4 in eqn.(1)] in their bubble summation procedure.
123: When converted into a physically more transparent valence and
124: conduction bond quasi-particle basis of our fermi liquid state, Peres
125: et al.'s susceptibility contains diagrams in which intermediate states
126: have higher quasi-particle numbers (an example is shown in figure 1).
127: Once we {\bf isolate and remove
128: these processes from their bubble summation}, we recover our fermi liquid
129: result (equation 2) and a full spin-1 collective mode branch emerges
130: from the bottom of the particle-hole continuum.
131:
132: \begin{references}
133:
134: \bibitem{peres} N.M.R. Peres, M.A.N. Araujo and A.H. Castro Neto,
135: preceding comment
136: \bibitem{gbak1} G. Baskaran and S.A. Jafari, Phys. Rev. Lett.
137: {\bf 89} 016402 (2002)
138: \bibitem{gbak2} G. Baskaran and S.A. Jafari, {\rm unpublished}
139: \bibitem{landau} L.D. Landau, Sov. Phys. JETP, {\bf 35} 70 (1959);
140: reprinted in `The Many-Body Problem' Ed., D. Pines (W.A. Benjamin inc.)
141: paragraph following eqn.(27)in page 277.
142: \end{references}
143: \end{multicols}
144: \end{document}
145: