1: \documentclass[twocolumn,prl,a4paper,aps,showpacs]{revtex4}
2: \usepackage{graphicx}
3: \usepackage{dcolumn}
4: \usepackage{bm}
5:
6: \begin{document}
7:
8: \title{Magnetic long-range order induced by quantum relaxation in single-molecule magnets}
9:
10: \author{M. Evangelisti$^{1}$, F. Luis$^{2}$, F. L. Mettes$^{1}$, N. Aliaga$^{3,*}$, G. Arom\'i$^{4}$, J. J. Alonso$^{5}$, G. Christou$^{3}$, and L. J. de Jongh$^{1}$}
11:
12: \affiliation{$^{1}$ Kamerlingh Onnes Laboratory, Leiden University, 2300 RA Leiden, The
13: Netherlands\\ $^{2}$ ICMA, CSIC and Universidad de Zaragoza, 50009 Zaragoza, Spain\\ $^{3}$
14: Department of Chemistry, University of Florida, FL 32611 Gainesville, USA\\ $^{4}$ Departament de
15: Qu\'imica Inorg\`{a}nica, Universitat de Barcelona, 08028 Barcelona, Spain\\ $^{5}$ Departamento
16: de F\'isica Aplicada I, Universidad de M\'alaga, 29071 M\'alaga, Spain}
17: \date{\today}
18:
19: \begin{abstract}
20: Can magnetic interactions between single-molecule magnets (SMMs) in a crystal establish long-range
21: magnetic order at low temperatures deep in the quantum regime, where the only electron
22: spin-fluctuations are due to incoherent magnetic quantum tunneling (MQT)? Put inversely: can MQT
23: provide the temperature dependent fluctuations needed to destroy the ordered state above some
24: finite $T_{c}$, although it should basically itself be a $T$-independent process? Our experiments
25: on two novel Mn$_{4}$ SMMs provide a positive answer to the above, showing at the same time that
26: MQT in the SMMs has to involve spin-lattice coupling at a relaxation rate equaling that predicted
27: and observed recently for nuclear spin-mediated quantum relaxation.
28: \end{abstract}
29:
30: \pacs{75.40.Cx, 75.45.+j, 75.50.Xx}
31:
32: \maketitle
33:
34: Despite the large number of studies on magnetic quantum tunneling (MQT) in molecular crystals of
35: single-molecule magnets (SMMs)~\cite{dante}, the question whether it is able to bring the spin
36: system into thermal equilibrium with the lattice, remains unsolved. Prokof'ev and
37: Stamp~\cite{fe8th} have suggested that interaction with rapidly fluctuating hyperfine fields can
38: bring a significant number of electron spins into resonance. Coupling to a nuclear spin bath
39: indeed allows ground-state tunneling over a range of local bias fields $\xi$ much larger than the
40: tunnel splitting $\Delta$, whereas, in its absence, tunneling would happen only if $\xi$ is
41: $\sim\Delta$ or less. Within this theory, magnetic relaxation could thus in principle occur with
42: no exchange of energy with the phonons of the molecular crystal~\cite{Fernandez03}. Support for
43: the Prokof'ev/Stamp model came from magnetic relaxation studies on the Fe$_{8}$ SMM~\cite{fe8exp}.
44: However, these experiments covered only initial stages of the relaxation process, leaving open the
45: question whether the final state corresponds to a thermal equilibrium. When MQT is combined with
46: coupling to a heat bath, dipolar couplings between cluster spins can induce long-range magnetic
47: ordering (LRMO) at sufficiently low temperatures~\cite{julio}. This phenomenon has not been
48: observed yet for any of the known SMMs relaxing by MQT. Attempts have been made in Fe$_{8}$ and
49: Mn$_{12}$ in particular, by increasing the tunneling rate by means of an applied transverse
50: field~\cite{trans}. However, the failure is inherent to the approach, since the applied fields
51: needed are much higher than the weak interaction energies involved, and will thus destroy the
52: LRMO.
53:
54: In this letter, we present zero-field time-dependent specific heat measurements performed on two
55: novel tetranuclear molecular clusters, both with net cluster spin $S=9/2$, denoted by Mn$_{4}$Cl
56: and Mn$_{4}$Me, which have similar cluster cores but different ligand molecules. For both, the
57: cluster spins become blocked along their anisotropy axes for temperatures in the liquid helium
58: range, and relaxation below $\sim 0.8$~K can thus only proceed by incoherent MQT between the two
59: lowest lying states $m=\pm S$. For both compounds we prove below that the MQT has to be inelastic.
60: For Mn$_{4}$Me, the tunneling rates are found sufficiently high to establish, even at zero field,
61: thermal equilibrium conditions down to the lowest temperatures ($\sim 0.1$~K). Accordingly, the
62: MQT channel enables the occurrence of LRMO between the cluster spins at $T_{c}=0.21$~K. Comparing
63: the magnitude of $T_{c}$ with Monte Carlo simulations suggests the coexistence of dipolar and weak
64: superexchange interactions between clusters. In view of the essential role of the dynamic nuclear
65: bias in the MQT mechanism, our results call for an extension of the nuclear-spin mediated quantum
66: relaxation model~\cite{fe8th} to include inelastic processes, where MQT is accompanied by phonon
67: creation or annihilation.
68:
69: Analytically pure samples of the compounds Mn$_{4}$O$_{3}$L(dbm)$_{3}$, with L~=~Cl(OAc)$_{3}$ or
70: [O$_{2}$C(C$_{6}$H$_{4}$-$p$-Me)]$_{4}$, hereafter abbreviated as Mn$_{4}$Cl and Mn$_{4}$Me, were
71: prepared as described in Ref.~\cite{mag}. All samples were characterized by elemental analysis.
72: Both molecules possess a distorted cubane core with one Mn$^{4+}$ ion (spin $S=3/2$) and three
73: Mn$^{3+}$ ions ($S=2$), superexchange coupled via three oxygen ions. The {\it intra}-cluster
74: exchange couplings have been studied by magnetic susceptibility measurements~\cite{mag}. Below
75: $T\lesssim 10$~K, the four Mn spins become ordered with a net total spin $S=9/2$ subject to an
76: uniaxial crystal field~\cite{uniaxial}, the symmetry axis running approximately through the
77: Mn$^{4+}$ ion and the L-ligand. Whereas Mn$_{4}$Cl has a local virtual $C_{3V}$ symmetry, the more
78: bulky carboxylate ligand distorts and lowers the symmetry ($C_{S}$) of Mn$_{4}$Me in the crystal
79: and increases the magnitude of the uniaxial anisotropy~\cite{mag,uniaxial}.
80:
81: Low-temperature specific heat measurements were performed by a home-made calorimeter using a
82: thermal relaxation method, see Ref.~\cite{trans}. By varying the thermal resistance of the thermal
83: link between calorimeter and cold-sink, the characteristic timescale $\tau_{e}$ of the experiment
84: can be varied. In this way time-dependent specific heat measurements can be exploited to
85: investigate spin-lattice relaxation when the relaxation time becomes of the order of $\tau_{e}$
86: ($\sim 1-100$~s)~\cite{cm}. The samples consisted of $1-3$~mg of polycrystalline material, mixed
87: with $2-5$~mg of Apiezon-N grease to ensure good thermal contact. The high-temperature ($2~{\rm
88: K}<T<300~{\rm K}$) specific heat was measured for a Mn$_{4}$Cl pellet sample of about $40$~mg
89: using a commercial (PPMS) calorimeter.
90:
91: The specific heat $C/R$ of Mn$_{4}$Cl is shown in Fig.~1, which combines data obtained for
92: $\tau_{e}\approx 2$~s, $8$~s, and $300$~s (as estimated at $T=0.4$~K) with the ones obtained with
93: the high-$T$ calorimeter. Let us first consider data measured for $T>1$~K, where $C$ is
94: independent of $\tau_{e}$. A $\lambda$-type anomaly is observed at $T\simeq 7$~K, having a
95: relative height of $2.5$~$R$. Measurements with an applied magnetic field (not presented here)
96: have shown that this anomaly is insensitive to the field, proving that it has to be associated to
97: a structural phase transition. Between $1$~K and $7$~K, $C$ is dominated by contributions from the
98: lattice phonons and from transitions between energy levels of the $S=9/2$ multiplet split by the
99: crystal field.
100:
101: Neglecting, in first approximation, the Zeeman splittings $\xi$ due to effective fields
102: representing the hyperfine interaction and the {\it inter}-cluster magnetic couplings, the
103: spectrum of energy levels is doubly degenerate at zero-applied-field, so that $C$ only depends on
104: transitions between levels inside each of the potential wells that are separated by the anisotropy
105: energy barrier $U$. We shall call this the {\it intra}-well contribution. The associated
106: multilevel Schottky anomaly is calculated with the eigenvalues of the spin Hamiltonian
107: \begin{equation}\label{Hamiltonian}
108: {\cal H}=-DS_{z}^{2}-E(S_{x}^{2}-S_{y}^{2})+A_{4}S_{z}^{4}
109: \end{equation}
110: with parameters $D$, $E$ and $A_{4}$ obtained independently from inelastic neutron scattering and
111: high-frequency EPR measurements~\cite{epr,parameters}. The {\it intra}-well specific heat
112: decreases exponentially as $T$ decreases and, since the first excited $m=\pm 7/2$ level is about
113: $(2S-1)D=5.5$~K above the $m=\pm 9/2$ ground-state doublet, it becomes almost negligible when $T
114: \leq 0.8$~K (Fig. 1). Adding the lattice contribution, calculated with a Debye temperature
115: $\theta_{D}\simeq 15$~K, to the Schottky accounts well for the experimental data (solid line in
116: Fig.~1). The lattice specific heat above 10~K appears to be composed of a number of Einstein-type
117: contributions (not shown).
118:
119: \begin{figure}[t!]
120: \centering{\includegraphics[angle=0,width=8cm]{fig1.eps}\caption{Temperature dependence of the
121: specific heat of Mn$_{4}$Cl measured for $\tau_{e}\approx 2$~s, 8~s and 300~s. Solid line, sum of
122: the Debye term of the lattice (dotted line) plus the Schottky contributions; dashed line,
123: calculated nuclear contribution $C_{nucl}$.}}
124: \end{figure}
125:
126: Below $1$~K, we expect the equilibrium magnetic specific heat ($C_{m}$) to be dominated by two
127: contributions. The first arises from incoherent MQT events inside the ground-state doublet that is
128: split by the action of the effective fields arising from hyperfine interactions and {\it
129: inter}-cluster dipolar coupling. The second is the specific heat $C_{nucl}$ of the nuclear spins
130: of Mn, whose energy levels are split by the hyperfine interaction with the atomic electron spins.
131: The dashed line in Fig.~1 represents $C_{nucl}$ calculated with the hyperfine constants
132: $A_{hf}=7.6$~mK and $A_{hf}=11.4$~mK for, resp., Mn$^{3+}$ and Mn$^{4+}$ ions, obtained from
133: ESR~\cite{chf}. Experiments performed for the longest $\tau_{e}\approx 300$~s show indeed a large
134: low-$T$ contribution. By contrast, the specific heat decreases by almost two orders of magnitude
135: when $\tau_{e}\approx 2$~s, evidencing that $\tau_{e}$ has a large effect in this temperature
136: range. This shows that the equilibrium between the relative populations of the $+9/2$ and $-9/2$
137: states cannot be established within $\tau_{e}$ if this is too short. We note that, for the
138: shortest $\tau_{e}$, the low-$T$ specific heat becomes even smaller than $C_{nucl}$, indicating
139: that both nuclear and electron spins are out of equilibrium. This is understandable, since the
140: only channel for the nuclear spins to exchange energy with the lattice is via the electron spins.
141: The strong connection between nuclear and electron spin-lattice relaxation has earlier been
142: observed for Mn$_{12}$, Mn$_{6}$ and Fe$_{8}$~\cite{trans,isotropic}.
143:
144: \begin{figure}[t!]
145: \centering{\includegraphics[angle=0,width=8cm]{fig2.eps}\caption{$T$-dependent specific heat of
146: Mn$_{4}$Me measured for $\tau_{e}\approx 4$~s. Dashed line, calculated nuclear contribution
147: $C_{nucl}$; solid line, sum of lattice (dotted line) plus Schottky contributions. Insets:
148: electronic $C_{m}$ and entropy variation $\Delta S$, after subtraction of $C_{nucl}$. Solid line,
149: calculated dipolar ordering for the easy axis along the $(110)$ direction (see text); dashed
150: lines, high-$T$ entropy limits for $S=1/2$ and $S=9/2$.}}
151: \end{figure}
152:
153: The experimental $C/R$ of Mn$_{4}$Me, measured for $\tau_{e}\approx 4$~s (as estimated at
154: $T=0.4$~K), is shown in Fig.~2 together with the calculated contributions of the lattice and the
155: nuclear spins, and the {\it intra}-well Schottky contribution. The experimental data display a
156: $\lambda$-anomaly at $T_{c}=0.21$~K that we attribute to the onset of LRMO. The Schottky anomaly
157: is calculated as for Mn$_{4}$Cl with parameters obtained from high-frequency
158: EPR~\cite{epr,parameters}. For $T>1$~K, the remaining contribution is given by the lattice and is
159: well described by the sum of a Debye term for the acoustic low-energy modes plus an Einstein term
160: for a higher energy mode, with values $\theta_{D}\simeq 12.3$~K and $\theta_{E}\simeq 22$~K for,
161: resp., the Debye and Einstein temperatures. Below $0.15$~K, the specific heat of Mn$_{4}$Me shows
162: a clear upturn that can be described by $C_{nucl}/R \simeq 4.27\times 10^{-3}/T^{2}$. This
163: contribution is well fitted using hyperfine constants $A_{hf}=8.7$~mK and $A_{hf}=13.8$~mK for,
164: resp., Mn$^{3+}$ and Mn$^{4+}$ ions. These values are of the same order as for Mn$_{4}$Cl and are
165: close to those reported by Zengh and Dismukes for a natural Mn$_{4}$ cluster~\cite{chf}.
166:
167: As seen in Fig.~2, the magnetic ordering peak and the calculated Schottky due to the splitting of
168: the $S=9/2$ multiplet are well separated in temperature. This is confirmed by the analysis of the
169: temperature dependence of the electronic entropy $\Delta
170: S(T)=\int^{\infty}_{0}[C_{m}(T)-C_{nucl}(T)]/T{\rm d}T$ (right inset of Fig.~2). As expected, the
171: total entropy of the electron spins tends to $R\ln(2S+1)$, with $S=9/2$, at high temperatures.
172: However, for the magnetic ordering region ($T<0.8$~K), $\Delta S$ corresponds to an effective spin
173: $S=1/2$, as appropriate for a two-level system. This proves that there only the two lowest levels
174: ($m=\pm 9/2$) are populated and contribute to $C_{m}$.
175:
176: We note that, although the anisotropy barrier $U\simeq DS^{2}-A_{4}S^{4}-|E|S^{2}$ amounts to
177: $\approx 14$~K for both Mn$_{4}$Me and Mn$_{4}$Cl, due to the lower symmetry of Mn$_{4}$Me the
178: 2nd-order off-diagonal term ($E$) in the Hamiltonian of Eq.~(\ref{Hamiltonian}), and thus also
179: $\Delta$ of the ground-state, will be much larger for this compound. This is confirmed by
180: high-frequency EPR experiments that give $E/D \simeq 0.21$, i.e., nearly $5$ times larger than for
181: Mn$_{4}$Cl \cite{epr,parameters}. We observe that this difference has a very large influence on
182: the spin-lattice relaxation. For similar $\tau_{e}$ values, the electron spins of the Mn$_{4}$Cl
183: molecule go off equilibrium below $T=0.8$~K (Fig.~1), whereas for Mn$_{4}$Me we observe thermal
184: equilibrium for electron and nuclear spins down to the lowest temperature. We conclude this from
185: the fact that ({\it i}) the total electronic entropy contribution equals the expected limit for
186: $S=9/2$; ({\it ii}) the remaining specific heat below 0.15~K agrees well with the calculated
187: $C_{nucl}$.
188:
189: As commonly found in molecular clusters, the magnetic core of Mn$_{4}$Me is surrounded by a shell
190: of non-magnetic ligand molecules. It follows that, {\it inter}-cluster superexchange interactions
191: are very weak, leaving the {\it inter}-cluster dipolar coupling responsible for magnetic
192: ordering~\cite{isotropic}. To check this for Mn$_{4}$Me, we have performed Monte Carlo
193: simulations, as described in Ref.~\cite{julio}, for a $S=9/2$ Ising model of magnetic dipoles
194: regularly arranged on the Mn$_{4}$Me lattice. We have repeated our calculations for several
195: orientations of the molecular easy axis (an example is given in the left inset of Fig.~2 for the
196: easy axis along the $(110)$ direction). The calculated $T_{c}$'s are always much smaller than the
197: experimental value. Consequently, the Mn$_{4}$Me molecules are also coupled by weak superexchange
198: interactions. Indeed, by adding an {\it inter}-cluster nearest-neighbor exchange interaction
199: $|J|/k_{B}\simeq 0.14$~K to our dipolar calculations, we reproduce the experimental $T_{c}$
200: value~\cite{juanjo}.
201:
202: In order to estimate the spin-lattice relaxation rate $\Gamma$ for Mn$_{4}$Cl at low temperatures,
203: we have used the relation for the time-dependent specific heat
204: $C_{m}(t)=C_{0}+(C_{eq}-C_{0})\left[1-e^{-\Gamma\tau_{e}}\right]$, where $C_{0}$ and $C_{eq}$ are
205: the adiabatic and equilibrium limits of the specific heat, respectively~\cite{cm}. For the
206: electron spins, $C_{0}$ is to good approximation given by the {\it intra}-well Schottky specific
207: heat, whereas the ``slow'' specific heat at equilibrium corresponds to excitations involving
208: transitions between the two wells. We have fitted, thus, the $C_{m}(\tau_{e})$ data of Mn$_{4}$Cl,
209: taking at each temperature for $C_{eq}$ the $C_{m}(T)$ measured for Mn$_{4}$Me (in the range
210: $T>T_{c}$). We show $\Gamma(T)$ in Fig.~3, together with data from ac-susceptibility and
211: (short-time) magnetic relaxation experiments~\cite{xmn4}. For $T>1.7$ K, $\Gamma$ follows the
212: Arrhenius law, with $U\simeq 13.5$~K and $\tau_{0}\approx 1.4\times 10^{-7}$~s. For $T\lesssim
213: 0.8$~K, $\Gamma$ deviates from this thermal-activation law, in remarkable agreement with the
214: magnetic relaxation data. Its weak temperature dependence confirms that relaxation to thermal
215: equilibrium is dominated by direct MQT transitions within the ground-state doublet.
216:
217: \begin{figure}[t!]
218: \centering{\includegraphics[angle=0,width=8cm]{fig3.eps}\caption{Spin-lattice relaxation rate of
219: Mn$_{4}$Cl: ($\ast$) and ($\bullet$)~\cite{xmn4} obtained from ac-susceptibility and magnetic
220: relaxation data; ($\circ$) obtained by fitting the $C_{m}(\tau_{e})$ data of Fig.~1 (see text).
221: Dashed line is fit of high-$T$ data to Arrhenius law; solid line is calculated for magnetic fields
222: $B_{x}=150$~G and $B_{z}=350$~G (see text).}}.
223: \end{figure}
224:
225: Summing up, we have proven that for both compounds thermal contact between spins and lattice is
226: established by MQT fluctuations. For Mn$_{4}$Me, the associated rate is even fast enough to
227: produce thermal equilibrium down to the lowest temperature and thus enable LRMO. This implies
228: $\Gamma\gtrsim 1/\tau_{e}\simeq 1$~s$^{-1}$. For Mn$_{4}$Cl, we find a lower rate
229: ($10^{-1}-10^{-3}$~s$^{-1}$) for $T\lesssim 0.8$~K. It is of interest to compare the experimental
230: $\Gamma$ for Mn$_{4}$Cl with predictions from conventional models for spin-lattice relaxation,
231: assuming that the $m=\pm 9/2$ energy levels of the cluster spins are time-independent. We have
232: simulated the effect of {\it inter}-cluster dipolar coupling and hyperfine interactions by
233: introducing static magnetic fields $B_{x}=150$~G and $B_{z}=350$~G~\cite{fields}. The presence of
234: these Zeeman terms is essential, otherwise tunneling would be forbidden for half-integer spin
235: $S=9/2$~\cite{halfodd}. We calculated $\Gamma$ by solving a master equation, including {\it
236: intra}- as well as {\it inter}-well transitions, induced by phonons only, between exact
237: eigenstates of the spin Hamiltonian of Eq.~(\ref{Hamiltonian}), as discussed in Ref.~\cite{cm} and
238: recently applied to the analysis of $C_{m}$ of Fe$_{8}$ and Mn$_{12}$ clusters~\cite{trans}. The
239: result (solid line in Fig.~3) agrees well with the activated behavior observed at high-$T$, but
240: fails to account for $\Gamma$ measured below $1$~K by six orders of magnitude! This large
241: discrepancy can not be ascribed to errors in the estimated elastic properties of the lattice. Both
242: the pre-factor of the Arrhenius law and the measured value of $\theta_{D}$ give a value of $c_{s}
243: \simeq 5(1)\times 10^{2}$ m/s for the speed of sound. By contrast, $\Gamma$ observed at low-$T$
244: would require $c_{s}$ and $\theta_{D}$ to be $15$ times smaller. This would give rise to a large
245: lattice specific heat well below $1$ K, which is not observed.
246:
247: It appears therefore that extension to {\it dynamic} hyperfine fields acting on the cluster-spin
248: levels, as proposed in Ref.~\cite{fe8th}, is indeed a necessary prerequisite for any model for the
249: MQT of SMMs. Such dynamic bias fields will sweep the tunneling levels with respect to one another,
250: thereby enabling incoherent Landau-Zener type tunneling events. The model predicts quantum
251: relaxation rates agreeing with experiments~\cite{fe8exp}, but so far the relaxation of the cluster
252: spins was thought to occur solely/primarily to the nuclear spin bath, relaxation to phonons was
253: only expected at much longer time-scales. Our specific heat experiments, in which obviously the
254: heat is transferred to the spins {\it via} the lattice, clearly demonstrate that in fact
255: spin-lattice relaxation {\it has to be} involved and at much the same fast rates! Since
256: application of `conventional' models for spin-lattice relaxation leads to rates orders of
257: magnitude too low, our data call for an extension of the Prokof'ev/Stamp model in which nuclear
258: spin-mediated MQT events are combined with creation or annihilation of phonons.
259:
260: The authors are indebted to J.F. Fern\'andez and P.C.E. Stamp for enlightening discussions, and
261: R.S. Edwards and S. Hill for HFEPR experiments on Mn$_{4}$Me. This work is part of the research
262: program of the Stichting voor Fundamenteel Onderzoek der Materie (FOM).
263:
264:
265: \begin{thebibliography}{99}
266:
267: \bibitem[*]{byline} Present address: Max-Planck-Institut f\"ur Strahlenchemie,
268: 45470 M\"ulheim an der Ruhr, Germany.
269: \bibitem{dante} See, for instance, D. Gatteschi and R. Sessoli, Angew.
270: Chem. Int. Edit. {\bf 42}, 268 (2003).
271: \bibitem{fe8th} N.V. Prokof'ev and P.C.E. Stamp, J. Low Temp. Phys. {\bf 104},
272: 143 (1996); Phys. Rev. Lett. {\bf 80}, 5794 (1998); Rep. Prog. Phys. {\bf 63}, 669 (2000).
273: \bibitem{Fernandez03} J.F. Fern\'andez and J.J. Alonso, Phys. Rev. Lett. {\bf
274: 91}, 047202 (2003).
275: \bibitem{fe8exp} W. Wernsdorfer {\it et al.}, Phys. Rev. Lett. {\bf 84}, 2965 (2000).
276: \bibitem{julio} J.F. Fern\'andez and J.J. Alonso, Phys. Rev. B {\bf 62}, 53 (2000);
277: J.F. Fern\'andez, {\it ibid.} {\bf 66}, 064423 (2002).
278: \bibitem{trans} F. Luis {\it et al.}, Phys. Rev. Lett. {\bf 85}, 4377 (2000);
279: F.L. Mettes, F. Luis, and L.J. de Jongh, Phys. Rev. B {\bf 64}, 174411 (2001).
280: %\bibitem{Fernandez02} J.F. Fern\'andez, Phys. Rev. B {\bf 66}, 064423 (2002).
281: \bibitem{mag} H. Andres {\it et al.}, J. Am. Chem. Soc. {\bf 50}, 12469 (2000);
282: G. Arom\'i {\it et al.}, Inorg. Chem. {\bf 41}, 805 (2002); N.
283: Aliaga, K. Folting, D.N. Hendrickson, and G. Christou, Polyhedron
284: {\bf 20}, 1273 (2001).
285: \bibitem{uniaxial} S. Wang {\it et al.}, Inorg. Chem. {\bf 35}, 7578 (1996).
286: \bibitem{cm} J.F. Fern\'andez, F. Luis, and J. Bartolom\'e, Phys. Rev. Lett. {\bf 80},
287: 5659 (1998).
288: \bibitem{epr} R.S. Edwards and S. Hill (private communication).
289: \bibitem{parameters} The parameters used in our calculations are $D=0.69$~K,
290: $E=-3.15\times 10^{-2}$~K, $A_{4}=-3.25\times 10^{-3}$~K for Mn$_{4}$Cl (from Ref.~\cite{mag}),
291: and $D=0.76$~K, $E=0.15$~K, $A_{4}=-4.3\times 10^{-3}$~K for Mn$_{4}$Me (from HFEPR~\cite{epr}).
292: \bibitem{chf} M. Zheng and G.C. Dismukes, Inorg. Chem. {\bf 35}, 3307 (1996).
293: \bibitem{isotropic} A. Bino {\it et al.}, Science {\bf 241}, 1479 (1988); A. Morello
294: {\it et al.}, Phys. Rev. Lett. {\bf 90}, 017206 (2003).
295: \bibitem{juanjo} For the exchange interaction between the magnetic dipoles, we consider
296: $\mathcal{H}_{ex}=2J\sum_{i\neq j}~S_{z}^{ (i)}S_{z}^{ (j)}$. Taking $S=1/2$ and $J\neq 0$ only
297: for nearest dipoles ($z=5$ in Mn$_{4}$Me), we calculate $T_{c}\simeq 0.2$~K for $|J|/k_{B}\simeq
298: 0.14$~K, irrespective of the sign of $J$.
299: \bibitem{xmn4} S.M.J. Aubin {\it et al.}, J. Am. Chem. Soc. {\bf 120}, 49981 (1998).
300: \bibitem{fields} Interaction with nuclear spins induces a distribution of bias $\xi$
301: of width $E_{0}\simeq N^{1/2}\hbar\omega_{0}/2$, where $N$ is the number of nuclear spins and
302: $\hbar\omega_{0}$ is the average hyperfine splitting~\cite{fe8th}. Off-diagonal terms of the
303: hyperfine interaction are of the same order of magnitude~\cite{chf}. In addition, {\it
304: inter}-cluster dipolar energies $E_{int}$ further broaden the distribution of bias. We roughly
305: estimate $E_{0}\approx 9\times 10^{-2}$~K and $E_{int}\approx k_{\rm{B}}T_{c}=0.2$~K, equivalent
306: to $B_{x}\simeq 150$~G and $B_{z}\simeq 350$~G, respectively.
307: \bibitem{halfodd} D. Loss, D.P. DiVincenzo, and G. Grinstein, Phys. Rev. Lett. {\bf 69},
308: 3232 (1992); J. von Delft and C.L. Henley, {\it ibid.} {\bf 69}, 3236 (1992).
309: \end{thebibliography}
310:
311: \end{document}
312: