Beyond switches: Ratcheting a particle energetically uphill with a compartmentalized molecular machine

Article abstract of DOI:10.1021/ja057664z

Here we correlate chemical (covalent), physical (thermodynamic), and statistical (population distribution) descriptions of behavior with the way that two new types of simple molecular machines (the threads of rotaxanes) perform the task of transporting a Brownian substrate (the rotaxane macrocycle) between two distinguishable binding sites. The first machine-substrate ensemble is a [2]rotaxane that operates through a mechanism that intrinsically causes it to change the average position of the macrocycle irreversibly. This contrasts with the behavior of classic stimuli-responsive molecular shuttles that act as reversible molecular switches. The second system is a compartmentalized molecular machine that is able to pump its substrate energetically uphill using the energy provided by a photon by means of an olefin photoisomerization. Resetting this compartmentalized molecular machine does not undo the work it has carried out or the task performed, a significant difference to a simple molecular switch and a characteristic we recognize as "ratcheting" (see Scheme 8). The ratcheting mechanism allows the [2]rotaxane to carry out the transport function envisaged for the historical thought-machines, Smoluchowski's Trapdoor and Maxwell's Pressure Demon, albeit via an unrelated mechanism and using an input of energy. We define and exemplify the terms "ratcheting" and "escapement" in mechanical terms for the molecular level and outline the fundamental phenomenological differences that exist between what constitutes a two-state Brownian switch, a two-state Brownian memory or "flip-flop", and a (two-stroke) Brownian motor. We also suggest that considering the relationship between the parts of a molecular machine and a substrate in terms of "statistical balance" and "linkage" could be useful in the design of more complex systems and in helping to understand the role of individual amino acids and peptide fragments during the directional transport of substrates by biological pumps and motors.

Full text of DOI:10.1021/ja057664z

Published on Web 03/02/2006
Beyond Switches: Ratcheting a Particle Energetically Uphill
with a Compartmentalized Molecular Machine
Manashi N. Chatterjee, Euan R. Kay, and David A. Leigh*
Contribution from the School of Chemistry, UniVersity of Edinburgh, The King’s Buildings,
West Mains Road, Edinburgh EH9 3JJ, United Kingdom
Received November 10, 2005; E-mail: David.Leigh@ed.ac.uk
Abstract: Here we correlate chemical (covalent), physical (thermodynamic), and statistical (population
distribution) descriptions of behavior with the way that two new types of simple molecular machines (the
threads of rotaxanes) perform the task of transporting a Brownian substrate (the rotaxane macrocycle)
between two distinguishable binding sites. The first machine-substrate ensemble is a [2]rotaxane that
operates through a mechanism that intrinsically causes it to change the average position of the macrocycle
irreversibly. This contrasts with the behavior of classic stimuli-responsive molecular shuttles that act as
reversible molecular switches. The second system is a compartmentalized molecular machine that is able
to pump its substrate energetically uphill using the energy provided by a photon by means of an olefin
photoisomerization. Resetting this compartmentalized molecular machine does not undo the work it has
carried out or the task performed, a significant difference to a simple molecular switch and a characteristic
we recognize as “ratcheting” (see Scheme 8). The ratcheting mechanism allows the [2]rotaxane to carry
out the transport function envisaged for the historical thought-machines, Smoluchowski’s Trapdoor and
Maxwell’s Pressure Demon, albeit via an unrelated mechanism and using an input of energy. We define
and exemplify the terms “ratcheting” and “escapement” in mechanical terms for the molecular level and
outline the fundamental phenomenological differences that exist between what constitutes a two-state
Brownian switch, a two-state Brownian memory or “flip-flop”, and a (two-stroke) Brownian motor. We also
suggest that considering the relationship between the parts of a molecular machine and a substrate in
terms of “statistical balance” and “linkage” could be useful in the design of more complex systems and in
helping to understand the role of individual amino acids and peptide fragments during the directional transport
of substrates by biological pumps and motors.
Introduction
we examine the way that some simple molecular machines carry
out the task of transporting a particle along a one-dimensional,
In recent years it has proved possible to design synthetic
molecular systems in which positional displacements of sub-
molecular components result from moving energetically down-
hill,1 but what are the structural features necessary for molecules
to convert chemical energy into mechanical work? How can
we make a synthetic molecular machine that pumps ions against
a gradient, say, or moves itself or a substrate energetically uphill
along a track? We know that nature has developed such
machines and refined them to a high degree of efficiency2 and
yet the chemistry literature is surprisingly poor when it comes
to the fundamental guidelines necessary to invent them. Here
two minimum, potential energy surface and attempt to correlate
the chemical (covalent structure), physical potential (thermo-
dynamic and kinetic properties governed by attractive and
repulsive noncovalent interactions), and statistical (population
distribution) behavior of the system with aspects of the task
performance. The results begin to provide the phenomenological
framework necessary for chemists to design more complex
compartmentalized molecular-level machines (assemblies of
simpler machines that each act as components by performing a
set task). It may also prove useful in understanding the roles
played by individual submolecular fragments during the opera-
tion of biological machines.
(1) (a) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. Angew. Chem.,
Int. Ed. 2000, 39, 3349-3391. (b) Stoddart, J. F., Ed. Special issue on
Molecular Machines. Acc. Chem. Res. 2001, 34, 409-522. (c) Balzani,
V.; Venturi, M.; Credi, A. Molecular DeVices and Machines. A Journey
into the Nanoworld; Wiley-VCH: Weinheim, Germany, 2003. (d) Easton,
C. J.; Lincoln, S. F.; Barr, L.; Onagi, H. Chem. Eur. J. 2004, 10, 3120-
3128. (e) Flood, A. H.; Ramirez, R. J. A.; Deng, W. Q.; Muller, R. P.;
Goddard, W. A.; Stoddart, J. F. Aust. J. Chem. 2004, 57, 301-322. (f)
Crowley, J. D.; Goshe, A. J.; Steele, I. M.; Bosnich, B. Chem. Eur. J.
2004, 10, 1944-1955. (g) Kottas, G. S.; Clarke, L. I.; Horinek, D.; Michl,
J. Chem. ReV. 2005, 105, 1281-1376. (h) Kinbara, K.; Aida, T. Chem.
ReV. 2005, 105, 1377-1400. (i) Kay, E. R.; Leigh, D. A. In Functional
Artificial Receptors; Schrader, T., Hamilton, A. D., Eds.; Wiley-VCH:
Weinheim, Germany, 2005; pp 333-406. (j) Kelly, T. R., Ed. Volume on
Molecular Machines. Top. Curr. Chem. 2005, 262, 1-236.
The Two-Compartment Brownian Particle
“Thought-Machines”
The design of tiny machines capable of transporting Brownian
particles selectively between two compartments;i.e., effectively
along a one-dimensional, two minimum, potential energy
surface;was the subject of several celebrated historical “thought-
machines” (Figure 1).3-7 Both Maxwell’s Demon4 (Figure 1a
and b) and Smoluchowski’s Trapdoor5 (Figure 1d) were
concerned with trying to set up temperature or pressure gradients
in systems containing multiple Brownian particles through their
(2) Molecular Motors; Schliwa, M., Ed.; Wiley-VCH: Weinheim, Germany,
2003.
9
4058
J. AM. CHEM. SOC. 2006, 128, 4058-4073
10.1021/ja057664z CCC: $33.50 © 2006 American Chemical Society

Compartmentalized Molecular Machines
A R T I C L E S
Figure 1. Examples of two-compartment Brownian “thought-machines”: (a) Maxwell’s “temperature demon” in which a gas at uniform temperature is
sorted into “hot” and “cold” molecules.4 Particles with energy higher than the average are represented by red dots while blue dots represent particles with
energies lower than the average. (b) A Maxwellian “pressure demon” in which a pressure gradient would be created if the door was only opened when a
particle in the left compartment approached it.4c (c) Szilard’s Engine, which attempts to do work with a piston using heat drawn from an external reservoir
by a pressure demon.6 (i) Initially, a single Brownian particle occupies a cylinder with a piston at either end. A frictionless partition is put in place to divide
the container into two compartments ((i) f (ii), unlinking stimulus). (ii) The demon then detects the particle and determines in which compartment it resides
(the left (L) compartment in the depicted example). (iii) Using this information, the demon is able to move the opposite piston into position without meeting
any resistance from the particle. (iv) The partition is removed (linking stimulus), allowing (v) the “gas” to expand against the piston, doing work against any
attached load. To replenish the energy used by the piston and maintain a constant temperature, heat must flow into the system. To complete the thermodynamic
cycle and reset the machine, the demon’s memory of where the particle was must be erased ((vi) f (i)). (d) Smoluchowski’s Trapdoor: an “automatic”
pressure demon. The directionally discriminating behavior is carried out by a wholly mechanical device, a trapdoor that is intended to open when hit from
one direction but not the other (note, this still involves the communication of information between the particle and the machine; the demon is incorporated
into the door mechanism).5
controlled exchange between two compartments; Szilard’s
Engine6 (Figure 1c) endeavored to utilize the pressure exerted
by one Brownian particle located in one of two compartments
to do work. The behaviors of all four “Gedankenmaschinen”
were considered without an external energy source (other than
a heat reservoir at the same temperature as the Gedankenm-
aschine system);their purpose was to test the nature of the
(3) Maxwell’s Demon 2. Entropy, classical and quantum information, comput-
ing; Leff, H. S., Rex, A. F., Eds.; Institute of Physics Publishing: Bristol,
U.K., 2003.
(4) The first (a) private and (b) public written discussions of the “temperature
demon” were as follows: (a) Maxwell, J. C. Letter to P. G. Tait, 11
December 1867. Quoted in Knott, C. G. Life and Scientific Work of Peter
Guthrie Tait; Cambridge University Press: London, 1911; pp 213-214.
(b) Maxwell, J. C. Theory of Heat; Longmans, Green and Co.: London,
1871; Chapter 12. (c) Maxwell introduced the idea of a “pressure demon”
in a later (undated) letter to Tait, also quoted in Knott, C. G. Life and
Scientific Work of Peter Guthrie Tait; Cambridge University Press: London,
1911; pp 214-215 and ref (3). A pressure demon is able to operate in a
system linked to a constant-temperature reservoir with the sole effect of
using energy transferred as heat from that reservoir to do work (see Szilard’s
engine, Figure 1c, ref 6). This is in conflict with the Kelvin-Planck form
of the Second Law, whereas the temperature demon challenges the Clausius
definition.
(8) For the realization of Feynman’s ratchet-and-pawl in molecular form see:
(a) Kelly, T. R.; Tellitu, I.; Sestelo, J. P. Angew. Chem., Int. Ed. Engl.
1997, 36, 1866-1868. For nonadiabatic molecular versions which unidi-
rectionally rotate see: (b) Kelly, T. R.; De Silva, H.; Silva, R. A. Nature
1999, 401, 150-152. (c) Leigh, D. A.; Wong, J. K. Y.; Dehez, F.; Zerbetto,
F. Nature 2003, 424, 174-179. (d) Herna´ndez, J. V.; Kay, E. R.; Leigh,
D. A. Science 2004, 306, 1532-1537. For other types of rotary molecular
motors see: (e) Koumura, N.; Zijlstra, R. W. J.; van Delden, R. A.; Harada,
N.; Feringa, B. L. Nature 1999, 401, 152-155. (f) Koumura, N.; Geertsema,
E. M.; Meetsma, A.; Feringa, B. L. J. Am. Chem. Soc. 2000, 122, 12005-
12006. (g) Koumura, N.; Geertsema, E. M.; van Gelder, M. B.; Meetsma,
A.; Feringa, B. L. J. Am. Chem. Soc. 2002, 124, 5037-5051. (h) Geertsema,
E. M.; Koumura, N.; ter Wiel, M. K. J.; Meetsma, A.; Feringa, B. L. Chem.
Commun. 2002, 2962-2963. (i) van Delden, R. A.; Koumura, N.; Harada,
N.; Feringa, B. L. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4945-4949. (j)
ter Wiel, M. K. J.; van Delden, R. A.; Meetsma, A.; Feringa, B. L. J. Am.
Chem. Soc. 2003, 125, 15076-15086. (k) van Delden, R. A.; Koumura,
N.; Schoevaars, A.; Meetsma, A.; Feringa, B. L. Org. Biomol. Chem. 2003,
1, 33-35. (l) ter Wiel, M. K. J.; van Delden, R. A.; Meetsma, A.; Feringa,
B. L. J. Am. Chem. Soc. 2005, 127, 14208-14222. (m) Pijper, D.; van
Delden, R. A.; Meetsma, A.; Feringa, B. L. J. Am. Chem. Soc. 2005, 127,
17612-17613. (n) ter Wiel, M. K. J.; van Delden, R. A.; Meetsma, A.;
Feringa, B. L. Org. Biomol. Chem. 2005, 3, 4071-4076. (o) van Delden,
R. A.; ter Wiel, M. K. J.; Pollard, M. M.; Vicario, J.; Koumura, N.; Feringa,
B. L. Nature 2005, 437, 1337-1340. (p) Fujita, T.; Kuwahara, S.; Harada,
N. Eur. J. Org. Chem. 2005, 4533-4543. (q) Kuwahara, S.; Fujita, T.;
Harada, N. Eur. J. Org. Chem. 2005, 4544-4556. (r) Fletcher, S. P.; Dumur,
F.; Pollard, M. M.; Feringa, B. L. Science 2005, 310, 80-82. (s) Lin, Y.;
Dahl, B. J.; Branchaud, B. P. Tetrahedron Lett. 2005, 46, 8359-8362.
(5) Smoluchowski’s Trapdoor [(a) von Smoluchowski, M. Phys. Z. 1912, 13,
1069-1080. (b) von Smoluchowski, M. Vortgage u¨ber die Kinetische
Theorie der Materie und der Elektrizitat; Planck, M., Ed.; Teubner und
Leipzig: Berlin, 1914; pp 89-121] aims to transport particles selectively
from the left compartment to the right in Figure 1d. However, in the absence
of a mechanism whereby the trapdoor can dissipate energy, it will be at
thermal equilibrium with its surroundings. This means it must spend much
of its time open, unable to influence particle transport. Rarely, it will be
closed when a particle approaches from the right and will open on collision
with a particle coming from the left;doing its job as intended. Such events
are balanced, however, by the door snapping shut on a particle from the
right, pushing it into the left chamber. Overall, the probability of a particle
moving from left to right is equal to that for moving right to left and so
the trapdoor cannot accomplish its intended function adiabatically.
(6) Szilard, L. Z. Phys. 1929, 53, 840-856.
(7) Feynman, R. P.; Leighton, R. B.; Sands, M. The Feynman Lectures on
Physics; Addison-Wesley: Reading, MA, 1963; Vol. 1, Chapter 46.
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4059

A R T I C L E S
Chatterjee et al.
Scheme 1. Previously Reported13j [2]Rotaxane, 1, that Functions as a Stimuli-Responsive Molecular Shuttlea
a At 254 nm the photostationary state is 61:39 E:Z,13j at 312 nm it is ∼50:50, and at 350 nm with a benzophenone sensitizer it is 40:60 E:Z13j and can
be increased up to 30:70 E:Z20j for some derivatives.
Second Law of Thermodynamics, not to see how a working
Brownian machine could be achieved (that was probably first
discussed7,8 by Feynman). However, modern synthetic routes
allow us to make molecules in which the Brownian motion of
substrates does occur between two well-defined locations, e.g.,
rotaxane-based molecular shuttles. This enables us to re-visit
the question of how to transport a Brownian particle between
two distinguishable sites, not from the point of view of doing
so adiabatically, but rather to see how such a task can be
performed by a molecular-level machine.
structure of one of the binding sites so as to change the relative
binding affinities of the stations for the macrocycle, placing the
system out of co-conformational19 equilibrium. Relaxation
toward the new global minimum subsequently occurs by the
macrocycle moving along the thread. A typical example13j of a
light-switchable amide-based molecular shuttle is shown in
Scheme 1.
However, we can also think of stimuli-responsive molecular
shuttles in another way; we can consider just the thread portion
of such a molecule as a machine that directionally transports a
particle;the interlocked macrocycle;between two sites (com-
partments) on a one-dimensional potential energy surface.20 The
significance of considering a rotaxane in this way is that over
Rotaxanes and Molecular Shuttles
Rotaxanes are chemical structures in which one or more
macrocycles are mechanically prevented from de-threading from
linear chains by bulky “stoppers”.9 Even though the rings are
not covalently attached to the threads, rotaxanes are molecules;
not supramolecular complexes;as covalent bonds must be
broken in order to separate the components from each other.9,10
The interactions generally used to direct the synthesis of
rotaxanes often “live-on” in the product, providing a well-
defined binding site or “station” for the ring on the thread. If
two or more stations are present on a thread with a traversable
path between them, the rotaxane can be considered a “molecular
shuttle”11 in which the ring is incessantly and randomly
exchanged between the binding sites.12 Stimuli-responsive
molecular shuttles are rotaxanes in which the net position of
the macrocycle on the thread (i.e., the statistical distribution of
the ring between the stations) changes in response to external
triggers (light,13 heat,14 electrons,15 chemical,16 pH,17 binding
events,18 etc.). Generally, the external stimulus alters the
(13) For examples of photochemically responsive molecular shuttles see: (a)
Benniston, A. C.; Harriman, A. Angew. Chem., Int. Ed. Engl. 1993, 32,
1459-1461. (b) Benniston, A. C.; Harriman, A.; Lynch, V. M. Tetrahedron
Lett. 1994, 35, 1473-1476. (c) Benniston, A. C.; Harriman, A.; Lynch, V.
M. J. Am. Chem. Soc. 1995, 117, 5275-5291. (d) Murakami, H.;
Kawabuchi, A.; Kotoo, K.; Kunitake, M.; Nakashima, N. J. Am. Chem.
Soc. 1997, 119, 7605-7606. (e) Armaroli, N.; Balzani, V.; Collin, J.-P.;
Gavin˜a, P.; Sauvage, J.-P.; Ventura, B. J. Am. Chem. Soc. 1999, 121, 4397-
4408. (f) Ashton, P. R.; Ballardini, R.; Balzani, V.; Credi, A.; Dress, K.
R.; Ishow, E.; Kleverlaan, C. J.; Kocian, O.; Preece, J. A.; Spencer, N.;
Stoddart, J. F.; Venturi, M.; Wenger, S. Chem. Eur. J. 2000, 6, 3558-
3574. (g) Wurpel, G. W. H.; Brouwer, A. M.; van Stokkum, I. H. M.;
Farran, A.; Leigh, D. A. J. Am. Chem. Soc. 2001, 123, 11327-11328. (h)
Brouwer, A. M.; Frochot, C.; Gatti, F. G.; Leigh, D. A.; Mottier, L.;
Paolucci, F.; Roffia, S.; Wurpel, G. W. H. Science 2001, 291, 2124-2128.
(i) Stanier, C. A.; Alderman, S. J.; Claridge, T. D. W.; Anderson, H. L.
Angew. Chem., Int. Ed. 2002, 41, 1769-1772. (j) Altieri, A.; Bottari, G.;
Dehez, F.; Leigh, D. A.; Wong, J. K. Y.; Zerbetto, F. Angew. Chem., Int.
Ed. 2003, 42, 2296-2300. (k) Abraham, W.; Grubert, L.; Grummt, U. W.;
Buck, K. Chem. Eur. J. 2004, 10, 3562-3568. (l) Murakami, H.;
Kawabuchi, A.; Matsumoto, R.; Ido, T.; Nakashima, N. J. Am. Chem. Soc.
2005, 127, 15891-15899. (m) Schmidt-Scha¨ffer, S.; Grubert, L.; Grummt,
U. W.; Buck, K.; Abraham, W. Eur. J. Org. Chem. 2006, 378-398. (n)
Wang, Q.-C.; Ma, X.; Qu, D.-H.; Tian, H. Chem. Eur. J. 2006, 12, 1088-
1096. (o) Balzani, V.; Clemente-Leo´n, M.; Credi, A.; Ferrer, B.; Venturi,
M.; Flood, A. H.; Stoddart, J. F. Proc. Natl. Acad. Sci. U.S.A. 2006, 103,
1178-1183.
(9) (a) Schill, G. Catenanes, Rotaxanes and Knots; Academic Press: New York,
1971. (b) Amabilino, D. B.; Stoddart, J. F. Chem. ReV. 1995, 95, 2725-
2828. (c) Molecular Catenanes Rotaxanes and Knots; Sauvage, J.-P.,
Dietrich-Buchecker, C. O., Eds.; Wiley-VCH: Weinheim, Germany, 1999.
(10) Hannam, J. S.; Lacy, S. M.; Leigh, D. A.; Saiz, C. G.; Slawin, A. M. Z.;
Stitchell, S. G. Angew. Chem., Int. Ed. 2004, 43, 3260-3264.
(11) (a) Anelli, P. L.; Spencer, N.; Stoddart, J. F. J. Am. Chem. Soc. 1991, 113,
5131-5133. For the first example of an amide-based molecular shuttle
see: (b) Lane, A. S.; Leigh, D. A.; Murphy, A. J. Am. Chem. Soc. 1997,
119, 11092-11093.
(12) The key feature of a rotaxane architecture from the point of view of
molecular-level machines is that movement of the macrocycle in any
direction other than along the thread is resisted by enormous steric forces
up until the breaking point of covalent bonds in the macrocycle or the
thread. This is a fundamentally different situation to a host-guest complex
when, following de-complexation from the binding site, the motion of the
ring is not restricted in any dimension and it is free to exchange with others
in the medium. Simple host-guest/supramolecular systems cannot function
as nanoscale mechanical machines unless restrictions on the exchange of
the unbound species with the bulk apply (as happens with kinetically stable
pseudo-rotaxanes) or the binding event brings about a mechanical (i.e.,
conformational) change in one of the molecular components. Similarly,
molecules that are not kinetically stable;this includes some rotaxanes that
are thermodynamically stable but kinetically labile;cannot behave as
molecular machines if they exchange components with the bulk quicker
than the time scale of their stimuli-induced change of position. Molecular
machines designed to exploit motion have to be kinetically associated with
their substrates throughout the operation of the machine.
(14) For an example of entropy-driven shuttling see: Bottari, G.; Dehez, F.;
Leigh, D. A.; Nash, P. J.; Pe´rez, E. M.; Wong, J. K. Y.; Zerbetto, F. Angew.
Chem., Int. Ed. 2003, 42, 5886-5889.
(15) For examples of electrochemically responsive molecular shuttles see ref
13e, 13o and (a) Bissell, R. A.; Co´rdova, E.; Kaifer, A. E.; Stoddart, J. F.
Nature 1994, 369, 133-136. (b) Collin, J.-P.; Gavin˜a, P.; Sauvage, J.-P.
New J. Chem. 1997, 525-528. (c) Ballardini, R.; Balzani, V.; Dehaen,
W.; Dell’Erba, A. E.; Raymo, F. M.; Stoddart, J. F.; Venturi, M. Eur. J.
Org. Chem. 2000, 591-602. (d) Ashton, P. R.; Ballardini, R.; Balzani, V.;
Credi, A.; Dress, K. R.; Ishow, E.; Kleverlaan, C. J.; Kocian, O.; Preece,
J. A.; Spencer, N.; Stoddart, J. F.; Venturi, M.; Wenger, S. Chem. Eur. J.
2000, 6, 3558-3574. (e) Altieri, A.; Gatti, F. G.; Kay, E. R.; Leigh, D. A.;
Martel, D.; Paolucci, F.; Slawin, A. M. Z.; Wong, J. K. Y. J. Am. Chem.
Soc. 2003, 125, 8644-8654. (f) Long, B.; Nikitin, K.; Fitzmaurice, D. J.
Am. Chem. Soc. 2003, 125, 15490-15498. (g) Kihara, N.; Hashimoto, M.;
Takata, T. Org. Lett. 2004, 6, 1693-1696. (h) Tseng, H.-R.; Vignon, S.
A.; Celestre, P. C.; Perkins, J.; Jeppesen, J. O.; Di Fabio, A.; Ballardini,
R.; Gandolfi, M. T.; Venturi, M.; Balzani, V.; Stoddart, J. F. Chem. Eur.
J. 2004, 10, 155-172. (i) Flood, A. H.; Peters, A. J.; Vignon, S. A.;
Steuerman, D. W.; Tseng, H.-R.; Kang, S.; Heath, J. R.; Stoddart, J. F.
Chem. Eur. J. 2004, 10, 6558-6564. (j) Tseng, H.-R.; Wu, D. M.; Fang,
N. X. L.; Zhang, X.; Stoddart, J. F. ChemPhysChem 2004, 5, 111-116.
(k) Steuerman, D. W.; Tseng, H.-R.; Peters, A. J.; Flood, A. H.; Jeppesen,
J. O.; Nielsen, K. A.; Stoddart, J. F.; Heath, J. R. Angew. Chem., Int. Ed.
2004, 43, 6486-6491. (l) Jeppesen, J. O.; Nygaard, S.; Vignon, S. A.;
Stoddart, J. F. Eur. J. Org. Chem. 2005, 196-220.
9
4060 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
the past decade physicists have developed formal theoretical
mechanisms, deeply rooted in nonequilibrium statistical me-
chanics, which explain how the directional transport of Brownian
particles can occur from periodic changes in a potential energy
surface (e.g., by applying an oscillating electric field).21,22 Many
different possible types of these theoretical “Brownian ratchet”
mechanisms have been suggested, including energy ratchets,
information ratchets, flashing ratchets, tilting ratchets, and
rocking ratchets.22 These mechanisms have been successfully
applied to the development of transport and separation devices
for mesoscopic particles and macromolecules, microfluidic
pumping, and quantum and electronic applications23,24 and have
also been shown to successfully account for the general
principles that govern the operation of complex biological
motors.24,25 However, to date it is not known how such
theoretical mechanisms correlate with the changes that occur
in molecular structure during the operation of biological
machines. What do individual peptide fragments do in order to
bring about transport of an ion or molecule by a Brownian
ratchet mechanism and why? We wondered whether examining
how these principles can be applied to some much less
sophisticated (in terms of function as well as structure) synthetic
molecular machines could tell us something about how they
might apply to more complex systems, both artificial and natural.
be considered to continuously fluctuate between the two stations.
However, even for a molecular shuttle with two different
stations, at equilibrium no net task can be performed by these
movements. This is a consequence of the “Principle of Detailed
Balance”;26 at equilibrium transitions between any two states
take place in either direction at the same rate so that no flux is
generated. This rules out the maintenance of equilibria by cyclic
(20) Stimuli-induced shuttling has been used to control a number of different
properties. See: Fluorescence switching: (a) Pe´rez, E. M.; Dryden, D. T.
F.; Leigh, D. A.; Teobaldi, G.; Zerbetto, F. J. Am. Chem. Soc. 2004, 126,
12210-12211. (b) Wang, Q.-C.; Qu, D.-H.; Ren, J.; Chen, K.; Tian, H.
Angew. Chem., Int. Ed. 2004, 43, 2661-2665. (c) Qu, D.-H.; Wang, Q.-
C.; Ren, J.; Tian, H. Org. Lett. 2004, 6, 2085-2088. (d) Qu, D.-H.; Wang,
Q.-C.; Tian, H. Mol. Cryst. Liq. Cryst. 2005, 430, 59-65. (e) Leigh, D.
A.; Morales, M. A. F.; Pe´rez, E. M.; Wong, J. K. Y.; Saiz, C. G.; Slawin,
A. M. Z.; Carmichael, A. J.; Haddleton, D. M.; Brouwer, A. M.; Buma,
W. J.; Wurpel, G. W. H.; Leo´n, S.; Zerbetto, F. Angew. Chem., Int. Ed.
2005, 44, 3062-3067. (f) Qu, D.-H.; Wang, Q.-C.; Tian, H. Angew. Chem.,
Int. Ed. 2005, 44, 5296-5299. (g) Qu, D.-H.; Wang, Q.-C.; Ma, X.; Tian,
H. Chem. Eur. J. 2005, 11, 5929-5937. (h) Li, Y.; Li, H.; Li, Y.; Liu, H.;
Wang, S.; He, X.; Wang, N.; Zhu, D. Org. Lett. 2005, 7, 4835-4838. (i)
Onagi, H.; Rebek, J. Chem. Commun. 2005, 4604-4606. Expression of
chirality: (j) Bottari, G.; Leigh, D. A.; Pe´rez, E. M. J. Am. Chem. Soc.
2003, 125, 13360-13361. Conductivity of solid-state electronic junctions:
(k) Collier, C. P.; Mattersteig, G.; Wong, E. W.; Luo, Y.; Beverly, K.;
Sampaio, J.; Raymo, F. M.; Stoddart, J. F.; Heath, J. R. Science 2000,
289, 1172-1175. (l) Collier, C. P.; Jeppesen, J. O.; Luo, Y.; Perkins, J.;
Wong, E. W.; Heath, J. R.; Stoddart, J. F. J. Am. Chem. Soc. 2001, 123,
12632-12641. (m) Luo, Y.; Collier, C. P.; Jeppesen, J. O.; Nielsen, K.
A.; Delonno, E.; Ho, G.; Perkins, J.; Tseng, H.-R.; Yamamoto, T.; Stoddart,
J. F.; Heath, J. R. ChemPhysChem 2002, 3, 519-525. (n) Diehl, M. R.;
Steuerman, D. W.; Tseng, H.-R.; Vignon, S. A.; Star, A.; Celestre, P. C.;
Stoddart, J. F.; Heath, J. R. ChemPhysChem 2003, 4, 1335-1339.
Conductivity at electrode interfaces: (o) Willner, I.; Pardo-Yissar, V.; Katz,
E.; Ranjit, K. T. J. Electroanal. Chem. 2001, 497, 172-177. (p) Sheeney-
Haj-Ichia, L.; Willner, I. J. Phys. Chem. B 2002, 106, 13094-13097. (q)
Katz, E.; Sheeney-Haj-Ichia, L.; Willner, I. Angew. Chem., Int. Ed. 2004,
43, 3292-3300. (r) Katz, E.; Lioubashevsky, O.; Willner, I. J. Am. Chem.
Soc. 2004, 126, 15520-15532. Exertion of a mechanical force on a
macroscopic object: (s) Huang, T. J.; Brough, B.; Ho, C.-M.; Liu, Y.; Flood,
A. H.; Bonvallet, P. A.; Tseng, H.-R.; Stoddart, J. F.; Baller, M.; Magonov,
S. Appl. Phys. Lett. 2004, 85, 5391-5393. (t) Liu, Y.; Flood, A. H.;
Bonvallet, P. A.; Vignon, S. A.; Northrop, B. H.; Tseng, H.-R.; Jeppesen,
J. O.; Huang, T. J.; Brough, B.; Baller, M.; Magonov, S.; Solares, S. D.;
Goddard, W. A.; Ho, C.-M.; Stoddart, J. F. J. Am. Chem. Soc. 2005, 127,
9745-9759. Access to a nanopore: (u) Nguyen, T. D.; Tseng, H.-R.;
Celestre, P. C.; Flood, A. H.; Liu, Y.; Stoddart, J. F.; Zink, J. I. Proc.
Natl. Acad. Sci. U.S.A. 2005, 102, 10029-10034. Surface wettability: ref
20r and (v) Berna´, J.; Leigh, D. A.; Lubomska, M.; Mendoza, S. M.; Pe´rez,
E. M.; Rudolf, P.; Teobaldi, G.; Zerbetto, F. Nat. Mater. 2005, 4, 704-
710. The latter also demonstrates the macroscopic transport of a droplet of
liquid across a surface using stimuli-responsive molecular shuttles.
(21) For an introduction to Brownian motors see: (a) Astumian, R. D. Sci. Am.
2001, 285 (1), 56-64. (b) Astumian, R. D.; Ha¨nggi, P. Phys. Today 2002,
55 (11), 33-39.
(22) For reviews of Brownian ratchet mechanisms see: (a) Ha¨nggi, P.; Bartussek,
R. In Nonlinear Physics of Complex Systems - Current Status and Future
Trends; Parisi, J., Mu¨ller, S. C., Zimmermann, W., Eds.; Lecture Notes in
Physics, Vol. 476; Springer: Berlin, 1996; pp 294-308. (b) Astumian, R.
D. Science 1997, 276, 917-922. (c) Ju¨licher, F.; Ajdari, A.; Prost, J. ReV.
Mod. Phys. 1997, 69, 1269-1281. (d) Special issue on The constructiVe
role of noise in fluctuation driVen transport and stochastic resonance. Chaos
1998, 8, 533-664. (e) Reimann, P. Phys. Rep. 2002, 361, 57-265. (f)
Reimann, P.; Ha¨nggi, P. Appl. Phys. A 2002, 75, 169-178. (g) Parrondo,
J. M. R.; De Cisneros, B. J. Appl. Phys. A 2002, 75, 179-191. (h) Gabrys,
B. J.; Pesz, K.; Bartkiewicz, S. J. Physica A 2004, 336, 112-122. (i) Linke,
H.; Downton, M. T.; Zuckermann, M. J. Chaos 2005, 15, 026111.
(23) (a) Rousselet, J.; Salome, L.; Ajdari, A.; Prost, J. Nature 1994, 370, 446-
448. (b) Faucheux, L. P.; Bourdieu, L. S.; Kaplan, P. D.; Libchaber, A. J.
Phys. ReV. Lett. 1995, 74, 1504-1507. (c) Bader, J. S.; Hammond, R. W.;
Henck, S. A.; Deem, M. W.; McDermott, G. A.; Bustillo, J. M.; Simpson,
J. W.; Mulhern, G. T.; Rothberg, J. M. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 13165-13169. (d) Linke, H.; Humphrey, T. E.; Lofgren, A.; Sushkov,
A. O.; Newbury, R.; Taylor, R. P.; Omling, P. Science 1999, 286, 2314-
2317. (e) Matthias, S.; Mu¨ller, F. Nature 2003, 424, 53-57.
Statistical Balance of a Dynamically Exchangeable
Substrate or Quantity (The Principle of Detailed
Balance)
If we ignore the normally small population of rings on the
spacer, at equilibrium the macrocycle in a molecular shuttle can
(16) For examples of chemically responsive molecular shuttles see ref 11b, 15h,
15l, and (a) Gong, C.; Gibson, H. W. Angew. Chem., Int. Ed. Engl. 1997,
36, 2331-2333. (b) Gong, C. G.; Glass, T. E.; Gibson, H. W. Macromol-
ecules 1998, 31, 308-313. (c) Jime´nez, M. C.; Dietrich-Buchecker, C.;
Sauvage, J.-P Angew. Chem., Int. Ed. 2000, 39, 3284-3287. (d) Lee, J.
W.; Kim, K.; Kim, K. Chem. Commun. 2001, 1042-1043. (e) Jimenez-
Molero, M. C.; Dietrich-Buchecker, C.; Sauvage, J.-P. Chem. Eur. J. 2002,
8, 1456-1466. (f) Da Ross, T.; Guldi, D. M.; Farran Morales, A.; Leigh,
D. A.; Prato, M.; Turco, R. Org. Lett. 2003, 5, 689-691. (g) Tseng, H.-
R.; Vignon, S. A.; Stoddart, J. F. Angew. Chem., Int. Ed. 2003, 42, 1491-
1495. (h) Laursen, B. W.; Nygaard, S.; Jeppesen, J. O.; Stoddart, J. F.
Org. Lett. 2004, 6, 4167-4170. (i) Huang, T. J.; Tseng, H.-R.; Sha, L.;
Lu, W. X.; Brough, B.; Flood, A. H.; Yu, B.-D.; Celestre, P. C.; Chang, J.
P.; Stoddart, J. F.; Ho, C.-M. Nano Lett. 2004, 4, 2065-2071. (j) Leigh,
D. A.; Pe´rez, E. M. Chem. Commun. 2004, 2262-2263. (k) Nørgaard, K.;
Laursen, B. W.; Nygaard, S.; Kjaer, K.; Tseng, H.-R.; Flood, A. H.;
Stoddart, J. F.; Bjørnholm, T. Angew. Chem., Int. Ed. 2005, 44, 7035-
7039.
(17) For examples of pH-responsive molecular shuttles see ref 15a and (a)
Mart´ınez-D´ıaz, M.-V.; Spencer, N.; Stoddart, J. F. Angew. Chem., Int. Ed.
Engl. 1997, 36, 1904-1907. (b) Ashton, P. R.; Ballardini, R.; Balzani, V.;
Baxter, I.; Credi, A.; Fyfe, M. C. T.; Gandolfi, M. T.; Go´mez-Lo´pez, M.;
Mart´ınez-D´ıaz, M.-V.; Piersanti, A.; Spencer, N.; Stoddart, J. F.; Venturi,
M.; White, A. J. P.; Williams, D. J. J. Am. Chem. Soc. 1998, 120, 11932-
11942. (c) Elizarov, A. M.; Chiu, S.-H.; Stoddart, J. F. J. Org. Chem. 2002,
67, 9175-9181. (d) Badjic´, J. D.; Balzani, V.; Credi, A.; Silvi, S.; Stoddart,
J. F. Science 2004, 303, 1845-1849. (e) Keaveney, C. M.; Leigh, D. A.
Angew. Chem., Int. Ed. 2004, 43, 1222-1224. (f) Garaude´e, S.; Silvi, S.;
Venturi, M.; Credi, A.; Flood, A. H.; Stoddart, J. F. ChemPhysChem. 2005,
6, 2145-2152. (g) Badjic´, J. D.; Ronconi, C. M.; Stoddart, J. F.; Balzani,
V.; Silvi, S.; Credi, A. J. Am. Chem. Soc. 2006, 128, 1489-1499.
(18) For examples of molecular shuttles switched by alkali metal ion complex-
ation, competitive binding, or allosteric regulation, see: (a) Vignon, S. A.;
Jarrosson, T.; Iijima, T.; Tseng, H.-R.; Sanders, J. K. M.; Stoddart, J. F. J.
Am. Chem. Soc. 2004, 126, 9884-9885. (b) Iijima, T.; Vignon, S. A.; Tseng,
H.-R.; Jarrosson, T.; Sanders, J. K. M.; Marchioni, F.; Venturi, M.; Apostoli,
E.; Balzani, V.; Stoddart, J. F. Chem. Eur. J. 2004, 10, 6375-6392. (c)
Marlin, D. S.; Gonza´lez Carbrera, D.; Leigh, D. A.; Slawin, A. M. Z. Angew.
Chem., Int. Ed. 2006, 45, 77-83. (d) Marlin, D. S.; Gonza´lez Carbrera,
D.; Leigh, D. A.; Slawin, A. M. Z. Angew. Chem., Int. Ed. 2006, 45, 1385-
1390.
(24) Linke, H., Ed. Special issue on Ratchets and Brownian motors: Basics,
experiments and applications. Appl. Phys. A 2002, 75, 167-352.
(25) (a) Astumian, R. D.; Bier, M. Biophys. J. 1996, 70, 637-653. (b) Astumian,
R. D.; Dere´nyi, I. Eur. Biophys. J. 1998, 27, 474-489. (c) Lipowsky, R.
In Stochastic Processes in Physics, Chemistry and Biology; Freund, J. A.,
Po¨schel, T., Eds.; Lecture Notes in Physics, Vol. 557; Springer: Berlin,
2000; pp 21-31. (d) Bustamante, C.; Keller, D.; Oster, G. Acc. Chem.
Res. 2001, 34, 412-420. (e) Astumian, R. D. Appl. Phys. A 2002, 75, 193-
206. (f) Mogilner, A.; Oster, G. Curr. Biol. 2003, 13, R721-R733. (g)
Oster, G.; Wang, H. Y. Trends Cell Biol. 2003, 13, 114-121. (h) Kurzynski,
M.; Chelminiak, P. Physica A 2004, 336, 123-132.
(19) “Co-conformation” refers to the relative positions of the mechanically
interlocked components with respect to each other, see: Fyfe, M. C. T.;
Glink, P. T.; Menzer, S.; Stoddart, J. F.; White, A. J. P.; Williams, D. J.
Angew. Chem., Int. Ed. Engl. 1997, 36, 2068-2070.
(26) Onsager, L. Phys. ReV. 1931, 37, 405-426.
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4061

A R T I C L E S
Scheme 2 a,b
Chatterjee et al.
a Description of E-1 and Z-1 in covalent, thermodynamic, and statistical terms, assuming a 90:10 distribution of translational isomers in each olefin
diastereomer. b“Machine-performance representation” of the forward (1A f 1C) and backward (1C f 1A) operations of machine-substrate system 1,
assuming a 50:50 olefin mixture at the photostationary state. The chemical structure of the machine (thread) is shown at each stage in cartoon form with the
fraction of olefin diastereomers indicated in terms of color of a station. The potential energy surface shown is the average potential energy surface of the
thread for the macrocycle at that stage of the machine operation, i.e., the appropriately weighted combination of the potential energy surfaces for each of the
olefin diastereomers (see (a)) contributing to the mixture. The population trace shows the overall distribution of the substrate over the machine.
processes such as A f B f C f A rather than A/B + B/C
+ C/A (ref 21b) and is a formal indication that a machine
such as Smoluchowski’s Trapdoor (Figure 1d) cannot operate
(at least not in the way originally envisaged). However, in
an out-of-equilibrium system, detailed balance is broken and
as the system moves spontaneously toward equilibrium net
work can be done by the fluxional exchange process. It is well-
established that breaking detailed balance is a requirement
for doing work with stochastic transport systems.22 How-
ever, we recently pointed out that detailed balance could
be considered to result from two separate properties of a
system;8d the statistical distribution of a quantity (an imbalance
in which provides the thermodynamic impetus for net transport)
and the ability of that quantity to be dynamically exchanged
(which provides the communication necessary for transport to
occur). It is no coincidence that all four of the Gedankenm-
aschinen shown in Figure 1 disconnect the compartments at
various stages during their operation in order to try and achieve
net particle transportation. The molecular examples in this paper
(vide infra) show that the deconvolution of the statistical
“balance” and the “linkage” of compartments is useful for
establishing the phenomenological nature of ratcheting and
escapement (the counterpart to ratcheting) in Brownian transport
processes.
9
4062 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
Scheme 3. A Type of [2]Rotaxane that can Act as an Irreversible Mechanical Switcha
a R3SiO- is a silyl ether that is too bulky to allow macrocycle exchange between the succinamide stations.
Now let us consider a new type of rotaxane system, 2, in
which a stimuli-induced change of position of the macrocycle
also occurs but through a clearly different mechanism to the
previous shuttle (Scheme 3). In 2 the two stations are structurally
identical (but distinguishable;note the differently depicted
stoppers) and a bulky group acts as a barrier which prevents
the ring from moving between them. If we start with 100% of
the rings on the first station, succ1-2,27 and then remove the
barrier, the system moves toward equilibrium, causing an
average displacement of the macrocycle of half the distance
separating the two stations.
Systematic Behavior of Some Simple Molecular
Machines
Let us consider how a stimuli-responsive rotaxane such as 1
(Scheme 1) performs the mechanical task of changing the
average position of the ring along the thread. So that we can
clearly see how the system evolves as this task is performed,
we will represent it at each stage (Scheme 2) in terms of the
covalent structure of the machine, that is, the thread (chemical
status), the average potential energy surface (physical status of
the machine, governed by noncovalent interactions;attractive
forces between machine and substrate such as H-bonding, and
repulsive forces such as steric barriers), and the statistical
distribution of the substrate being transported (the macrocycle).
In these types of schemes, which we shall call “machine-
performance representations”, we will use bold numbers to show
different systems (1, 2, 3... etc.) and lettered suffixes to
differentiate states of the system (1A, 1B, 1C.... etc.). In Scheme
2 we assume the photostationary state is 50:50 E:Z and that the
preferred ratio of occupancy between fumaramide (green) and
succinamide (orange) stations is 90:10 and between maleamide
(dark blue) and succinamide 10:90.
Again, we can see how this system evolves more clearly using
a “machine-performance representation” (Scheme 4). Unlike
system 1, machine-substrate system 2 starts out statistically
unbalanced (2A). The stimulus required is also different from
that used in the first system: a “linking stimulus” lowers the
barrier between the two stations (2B), allowing the system to
fulfill the impetus to restore balance and move to equilibrium
by biased Brownian motion of the macrocycle (2C, note that
the energy well-depths of the two stations are the same; the
macrocycle is not held more tightly by the station it moves to).
Raising the barrier by applying an “unlinking stimulus” resets
the machine;the thread;but this time the task it has performed
is not undone (2D). However, we note that if we try applying
the linking stimulus again after the machine is reset, the machine
does not change the average position of the macrocycle because
the system is already statistically balanced (Scheme 4).
The initial state of the system (1A) is statistically balanced.
The average position of the macrocycle (from the Boltzmann
distribution) is very close to the green station. A photonic
stimulus (“balance-breaking stimulus”) causes isomerization of
some of the olefins from E to Z, putting the position of the
macrocycle in the molecules that are isomerized momentarily
out of equilibrium (1B, similarly 1D). The system acts to restore
balance through biased Brownian motion of the macrocycles
and we arrive at the final state, 1C. The change in state between
1B and 1C (and also between 1D and 1A) is not triggered
externally; rather it is a thermally activated relaxation step in
which the average position of the macrocycle (given by the
Boltzmann distributions within the olefin isomers multiplied by
their contribution to the photostationary state) moves along the
thread toward the orange station. Note, however, that the
machine;the thread;cannot use the energy from the photon
to perform the transportation task in such a way that the position
of the substrate becomes independent of the state of the machine;
applying a second stimulus to reset the machine (e.g., adding
piperidine to reform the E-olefin of the thread) undoes the net
displacement of the macrocycle (1C f 1A).
This stimuli-induced irreversible net change of position of
the macrocycle represents a new type of molecular shuttle in
phenomenological terms and so we prepared an experimental
example, 3, in single translational isomer form (succ1-3)
according to Scheme 5. The unoccupied thread, 4, was also
synthesized as a comparison for 1H NMR purposes. To be able
to distinguish between the two succinamide stations through
1H NMR spectroscopy but still keep them similar in terms of
macrocycle binding affinity, the terminal right-hand side (as
depicted in Scheme 5) amide group was derivatized with a
(27) The nomenclature used to denote individual rotaxane configurational and
translational isomers follows conventions introduced in previous papers
(e.g., ref 8d). Namely, a bolded compound number refers to a given
structural formula irrespective of functional group configuration or position
of the macrocycle; a prefix E or Z describes the configuration of the olefin;
a prefix succ1, succ2, fum, or mal denotes the position of the macrocycle
in a particular translational isomer.
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4063

A R T I C L E S
Scheme 4 a,b
Chatterjee et al.
a Description of succ1-2 in covalent, thermodynamic, and statistical terms. b “Machine-performance representation” of a [2]rotaxane that acts as an irreversible
mechanical switch. 2A: The macrocycle is locked on one of the two energetically identical but distinguishable stations by a large kinetic energy barrier. 2B:
A linking stimulus lowers the barrier. 2C: The thermal bath restores equilibrium via biased Brownian motion. 2D: An unlinking stimulus resets the machine.
Note that repeating the cycle of chemical reactions would not change the distribution of the macrocycle for a second time.
phenyl group. Aryl-substituted tertiary amides preferentially
The synthetic route to succ1-3 is worthy of some comment.
The “gated” spacer that separates the two stations was prepared
by double alkylation (orthogonal amine protecting groups) and
subsequent Krapcho decarboxylation31 of dimethyl malonate
(Scheme 5, i-v). A fumaramide group was then attached, both
to maximize the yield of rotaxane formation (step viii, 68%)
and to provide a common route for the synthesis of fum-E-5
(vide infra). Unmasking of the Fmoc-protected amine, followed
by coupling with a succinic acid derivative (step ix), was
adopt an anti-arrangement of the alkyl groups so there is no
1
complication with slowly interconverting rotamers in the H
NMR spectra.28,29 (Tertiary aryl-substituted fumaramide groups
are excellent templates for benzylic amide macrocycle rotaxane
formation.30) A 4-picolyl fragment provides the necessary bulk
for the rest of the stopper.
(28) Itai, A.; Toriumi, Y.; Saito, S.; Kagechika, H.; Shudo, K. J. Am. Chem.
Soc. 1992, 114, 10649-10650.
(29) Note that there is no indication of multiple tertiary amide rotamers in the
1H NMR spectrum of 4 (Figure 2a). For example, only one signal is
observed for Hq.
(30) Leigh, D. A.; Nash, P. J. Manuscript in preparation.
(31) (a) Krapcho, A. P. Synthesis 1982, 805-822. (b) Krapcho, A. P. Synthesis
1982, 893-914.
9
4064 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
Scheme 6. Chemical Representation of the Irreversible Net
Scheme 5. Synthesis of Single Translational Isomer [2]Rotaxanes
succ1-3 and fum-E-5 and the Corresponding Threads, 4 and E-6a
Translocation of the Macrocycle that Occurs through the
Transformation succ1-3 f [succ2-3 + succ1-3] (50:50 ( 2)a
a Reaction conditions: (i) 80% aqueous acetic acid, 60 °C, 1 h. (ii)
TBDMSCl, imidazole, DMAP, CH2Cl2, rt, 1 h.
E-1 for example (Scheme 1), the translational isomers of 3 are
actually diastereoisomers.10 The 1H NMR spectra of the machine
in the absence of the substrate (i.e., the free thread, Figure 2a),29
the machine-substrate ensemble in its initial state (Figure 2b),
and after application of the linking and unlinking stimuli (Figure
2c) are shown in Figure 2.
a Reaction conditions (unless otherwise stated, reactions were carried
out at room temperature): (i) a. NaH, THF, 0 °C to rt, 1 h; b. Bu4NI,
N-benzyloxycarbonyl-5-bromo-1-pentylamine, THF, reflux, 12 h, 88%. (ii)
a. NaH, THF, 0 °C to rt, 1 h; b. Bu4NI, N-allyloxycarbonyl-5-bromo-1-
pentylamine, THF, reflux, 12 h, 58%. (iii) LiCl, DMSO, H2O, 160 °C, 4 h,
66%. (iv) Diisobutylaluminum hydride (1 M in toluene), THF, -78 °C, 4
h, 68%. (v) a. PhSiH3, Pd(PPh3)4, CH2Cl2, 45 min; b. fluorenylmethylchlo-
roformate (Fmoc-Cl), Et3N, 0 °C, 30 min, 56%. (vi) a. H2, 10% Pd/ C,
HCl (1 M in Et2O), MeOH, 1 atm, 1 h; b. (E)-3-(2,2-diphenylethylcarbam-
oyl)-acrylic acid, 1-[3-dimethylaminopropyl]-3-ethylcarbodiimide hydro-
chloride (EDCI‚HCl), 1-hydroxybenzotriazole hydrate (HOBt‚H2O), Et3N,
CH2Cl2, 0 °C to rt, 2 h, 60%. (vii) tert-Butyldimethylsilyl chloride
(TBDMSCl), imidazole, 4-(dimethylamino)pyridine (DMAP), CH2Cl2, 1
h, 63%. (viii) p-Xylylenediamine, isophthaloyl dichloride, Et3N, CHCl3, 3
h, 68%. (ix) a. Piperidine, THF:CH3CN (1:3), 2.5 h; b. 4-oxo-4-(phenyl(py-
ridin-4-ylmethyl)amino)butanoic acid, EDCI‚HCl, HOBt‚H2O, Et3N, CH2Cl2,
0 °C to rt, 14 h, 47% (E-6), 48% (fum-E-5). (x) H2, 10% Pd/C, THF, 1
atm, 6 h, 90% (4), 90% (succ1-3). Full experimental procedures can be
found in the Supporting Information.
Again, let us consider what the mechanism of operation of
the machine is: The stimuli applied to 3 switch “on” and “off”
whether the stations are able to exchange the macrocycle or
not. Linking of two parts of the machine which interact with
the substrate allows the system to move toward equilibrium,
that is, toward a statistically balanced state. This is a molecular-
level form of “escapement”, the element of the mechanism that
controls the release of potential energy to drive mechanical
motion in clocks and other macroscopic mechanical devices.34
Let us combine the features of the first two rotaxanes, 1 and
3, to invent a third type of molecular machine system, 5, and
see how it functions. Rotaxane 5 was synthesized as a single
configurational and translational isomer, fum-E-5, along with
the corresponding thread E-6, according to Scheme 5. The
reaction profile of this rotaxane is outlined in Scheme 7 in
standard chemical terms and its performance as a machine-
substrate ensemble is shown in Scheme 8.
followed by hydrogenation (step x) of the fumaramide moiety.
Perhaps surprisingly, the macrocycle appears to offer no
significant protection to the thread functionality in this step,32
which readily affords the two station rotaxane, 3, as a single
translational isomer, succ1-3 (however, careful solvent selection
is necessary to avoid cleavage of the silyl ether33).
The system functions (Scheme 6) as predicted in Scheme 4.
De-silylation of succ1-3 (Scheme 6, step i) followed by
re-silylation (step ii) afforded a ∼1:1 mixture of the two
translational isomers of 3. Note, because succ1-3 and succ2-3
are not in equilibrium with each other, unlike fum-E-1 and succ-
The xylylene rings of the macrocycle shield the regions of
the thread that they encapsulate and the resulting shifts in the
1H NMR spectra are diagnostic of the position of the macrocycle
on the thread.13j Thus, the operation of the machine and its effect
1
on the substrate can be followed by H NMR spectroscopy
(Figure 3). In particular, it is instructive to observe the change
in intensity of the unoccupied fumaramide station protons Hd′
and He′ (Hd and He in E-6) during the operation of the machine.
In the free thread they appear as a pair of doublets at 6.86 and
(32) Parham, A. H.; Windisch, B.; Vo¨gtle, F. Eur. J. Org. Chem. 1999, 1233-
1238.
(34) (a) Britten, F. J. Escapements: Their Actions Constructions and Proportion;
Arlington Book Co: U.S.A., 1985. (b) Gazeley, W. J. Clock and Watch
Escapements; Robert Hale & Co.: 1999.
(33) Sajiki, H.; Ikawa, T.; Hattori, K.; Hirota, K. Chem. Commun. 2003, 654-
655.
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4065

A R T I C L E S
Chatterjee et al.
Figure 2. 1H NMR spectra (400 MHz, CDCl3, 298 K) of (a) free thread 4, (b) succ1-3, and (c) mixture of translational isomers succ1-3 and succ2-3 after
1
the two-step operation (Scheme 6, (i) and (ii)). Integration shows the succ1-3:succ2-3 ratio in (c) is 50:50 ((2%). The H NMR assignments and coloring
correspond to the labeling in Schemes 5 and 6. Residual water peaks are shown in gray.
7.02 ppm (Figure 3a); they are absent in the spectrum of pure
fum-E-5 (Figure 3b); in the statistically balanced (unlinked)
system (Figure 3c) they account for ∼15% of the overall
population; in the final ratcheted system (Figure 3d) they are
∼56% of the reaction mixture.
the macrocycle is 85:15 between the fum and succ stations of
E-5 (Figure 3c), the final 44:56 fum-E-5:succ-E-5 ratio indicates
that the equilibrium distribution of translational isomers between
the mal and succ stations in the de-silylated (linked) derivative
of Z-5 is ∼5:95 in CH2Cl2 at room temperature.37 The
transportation of the macrocycle in 5 is repeatedly reversible
between the statistically balanced 85:15 and statistically unbal-
anced 44:56 ratios of fum-E-5 to succ-E-5.38
So, in Scheme 8, the two parts of the machine start out (5A)
statistically balanced (85% of the macrocycles on the fumara-
mide station; 15% on the succinamide station) and unlinked
(and therefore not in equilibrium). A balance-breaking stimulus
(hV at 312 nm,35 which generates a 49:51 (2% E:Z photosta-
The thread has therefore successfully performed the task of
directionally changing the net position of the macrocycle;and
since the succinamide station binds the macrocycle more weakly
than the fumaramide station, the thread has moved the macro-
cycle energetically uphill!;and the machine, the thread, has
returned to its initial state. We recognize this behavior as
“ratcheting”, a characteristic of the operating mechanism of
many biological molecular machines. For the first time, we have
a molecular shuttle which is more sophisticated in terms of task
performance than a simple mechanical switch. The compart-
mentalized machine achieves this result by applying four
different stimuli which govern in turn the thermodynamics and
the kinetics for transport between the two stations: balance-
breaking 1; linking; unlinking; balance-breaking 2 (resetting;of
the machine, not the substrate).39
1
tionary state by H NMR, not shown) is applied, giving 5B.
Removal of the barrier (“linking stimulus”; 80% aqueous acetic
acid) gives 5C and allows balance to be restored through moving
to equilibrium by biased Brownian motion of the ring (5D).
Restoring the barrier (“unlinking stimulus”; TBDMSCl, base)
makes the system (5E) unlinked and not in equilibrium, although
statistically balanced. The resetting step (a different balance-
breaking stimulus; catalytic piperidine, to promote the Z f E
olefin isomerization) makes the system statistically unbalanced,
unlinked, and not in equilibrium (5F).
After the operational cycle of the machine (i.e., 5F) the
unoccupied fumaramide station protons Hd′ and He′ account for
∼56% of the reaction mixture (Figure 3d).36 Given that the
photostationary state from irradiation of E-5 at 312 nm is 49:
51 (2% E:Z, and that the statistically balanced distribution of
(37) This calculation assumes that both translational isomers of de-silylated Z-5
are re-silylated at the same rate. If this is not true, if the position of the
macrocycle on a station affects the rate of the unlinking reaction, then it is
possible to produce a nonbalanced distribution of the macrocycle through
just controlling kinetic energy barriers with the position of the ring;an
“information ratchet” (ref 25b).
(38) The pure single translational isomer fum-E-5 cannot be regenerated by the
operation of the machine but it can be separated from the reaction mixture
at any stage using standard purification protocols.
(35) Some degradation of rotaxane 5 occurs upon irradiation at 254 nm; however,
at 312 nm no degradation or side reactions are apparent and the
photostationary state is 49:51 ( 2 E:Z by 1H NMR spectroscopy.
(36) Note how the chemical shifts of some of the protons (e.g., Hd and He) in
fum-E-5 change among Figures 3b, 3c, and 3d, illustrating the sensitivity
of the H-bonding between the macrocycle and thread to factors such as
moisture, concentration, and possibly, impurities.
9
4066 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
Scheme 7. Chemical Representation of the Operation of Machine-Substrate System 5a
a Reaction conditions: (i) 80% aqueous acetic acid, 60 °C, 1 h. (ii) hV at 312 nm (5 × 5 min irradiation),35 CH2Cl2, rt. (iii) TBDMSCl, imidazole, DMAP,
CH2Cl2, rt, 1 h. (iv) Piperidine, CH2Cl2, rt, 12 h, quantitative.
Through its operation in Scheme 7/8, rotaxane 5 estab-
lishes a thermodynamically unfavorable concentration grad-
ient of the macrocycle between the compartments. This is
the function envisaged for the thought-machine pressure de-
mons shown in Figures 1b and 1d, albeit featuring many
machines each acting on one Brownian particle in the case of
the rotaxane rather than a single machine acting on many
Brownian particles. However, the way in which 5 achieves
this result is very different from either Gedankenmaschine
design. During the operation of 5 the Brownian particle’s
position does not determine when or whether the linking
stimulus is applied. This would require the communication of
information regarding the particle’s position to the machine,
which is possible but would correspond to an “information
ratchet”25b mechanism. Rather the rotaxane machine carries out
its operations independent of the position of the particle by
varying the potential energy surface minima as well as maxima
in a partial so-called “energy ratchet”25b,8d mechanism (Figure
4).
As mechanical work is done by its operation, is it correct to
categorize 5 as a molecular motor? No. Although the machine
component of 5 acts to transport a substrate energetically uphill
and can be reset without undoing the work done on the substrate,
it cannot do so repetitively;a key requirement of a motor;be-
cause the succinamide compartment cannot be emptied of the
ratcheted substrate. The escapement of the ratcheted quantity
is the missing element required for 5 to operate for a second
time without undoing the previously performed task by the
action of resetting the machine.40 This feature is present in a
previously reported8d [2]catenane, 7, in which the larger
macrocycle acts as a motor that repetitively transports the small
macrocycle directionally around itself according to a cyclic
reaction scheme (Scheme 9). As with 5, the olefin isomerization
reactions in Scheme 9 are balance-breaking steps (depending
on where one starts in the scheme, either can be considered to
reset the machine); the manipulation of the trityl and silyl
protecting groups are pairs of linking-unlinking steps which
determine the pathway through which escapement of the
ratcheted substrate occurs. Thus, the molecular machine in
Scheme 9 operates by the following sequence: [balance-
breaking 1; escapement (pathway A); ratcheting; balance-
breaking 2 (reset machine); escapement (pathway B); ratchet-
ing]n.
(39) Starting from state 5A, the linking and initial balance-breaking stimuli can
be applied in either order (or simultaneously) in Scheme 8 for the task to
be performed by the machine on the substrate, i.e., net transport of the
macrocycle along the thread. In fact, any orthogonal transformations that
transpire between two unlinking operations are commutative. Indeed, in a
classic energy ratchet mechanism (ref 25b);say using an oscillating electric
field to directionally transport a charged particle;the linking/unlinking
and balance-breaking steps happen simultaneously which, of course, would
have exactly the same effect as doing them sequentially as envisaged in
Scheme 8. Balance-breaking and linking/unlinking steps also occur
simultaneously during the operation of molecular shuttles containing
cyclodextrins incorporated onto azobenzene or stilbene threads (refs 13d,
13i, 13l, 13n, 20b, 20c, 20d, 20f, and 20g).
(40) Although escapement occurs during the operation of 5, it is not escapement
of a quantity that had been previously ratcheted by the operation of the
motor. Prior to escapement, the macrocycles located on the olefin station
in 5 have only been statistically unbalanced by the action of the motor, not
ratcheted.
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4067

A R T I C L E S
Scheme 8 a,b
Chatterjee et al.
a Description of fum-E-5 and mal-Z-5 in covalent, thermodynamic, and statistical terms. b“Machine-performance representation” of a compartmentalized
molecular machine that can transport a substrate energetically uphill and be reset without undoing the task. 5A: Initially balanced and unlinked. 5B: Unbalanced
and unlinked. 5C: Unbalanced and linked (detailed balance is broken). 5D: Balanced and linked. 5E: Balanced and unlinked. 5F: Unbalanced and unlinked.
Note the chemical structure of the thread in 5F is identical to that in 5A; the machine has been reset;but the population distribution of the macrocycle
has changed. The machine has successfully utilized the energy of the photon, via the olefin isomerization reactions, to do the work required to transport the
substrate energetically uphill. The stimuli correspond to the reaction conditions given in Scheme 7.
Although rotaxane 5 does not fulfill the requirements for a
motor, neither is it a simple switch since the machine part can
be reset without influencing the distribution of the substrate.
Significantly, examining the state of the machine (the rotaxane
thread) in 5 does not provide information regarding the state
(distribution) of the substrate. It is only from the history of the
9
4068 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
Figure 3. 1H NMR spectra (400 MHz, CDCl3, 298 K) of (a) free thread E-6, (b) fum-E-5, (c) statistically balanced E-5, the 85:15 mixture of fum-E-5 and
succ-E-5 that results from desilylation/resilylation of fum-E-5 (Scheme 7, (i) and (iii)), and (d) 44:56 ((2%) mixture of fum-E-5 and succ-E-5 that results
from the four-step operation of the machine, either starting from fum-E-5 or the 85:15 fum-E-5:succ-E-5 statistically balanced mixture (Scheme 7, (i), (ii),
(iii), (iv) or (ii), (i), (iii), (iv); Scheme 8, 5A f 5F). The 1H NMR assignments and coloring correspond to the labeling in Schemes 5 and 7. Residual water
peaks are shown in gray.
Figure 4. The operation of machine-substrate 5 in Schemes 7 or 8 is the experimental realization (albeit nonadiabatic) of the transportation task required
of Smoluchowski’s Trapdoor5 (Figure 1d) and Maxwell’s Pressure Demon4c (Figure 1b). The mechanistic equivalent of the illustrations from Figure 1 is
shown above. The colors of the compartments, particles, and door are the same as the corresponding elements of 5. The initially balanced (in proportion to
the sizes of the two compartments) distribution of the Brownian particles between the left (L) and right (R) compartments becomes statistically unbalanced
by a change in volume of the left-hand compartment. Opening the door allows the particles to redistribute themselves according to the new size ratio of the
compartments. Closing the door ratchets the new distribution of particles. Restoring the left-hand compartment to its original size then results in a concentration
gradient of the Brownian particles across the two compartments. There is no role for an information-gathering demon in this mechanism.
machine’s operations that the distribution of the substrate can
be known; in other words, the net position of the macrocycle
in rotaxane 5 is a consequence of a form of sequential logic.
This type of logic is different from that utilized in most of the
Boolean logic chemical systems investigated to date, which
feature combinational logic (the outputs are solely a function
of the inputs at that moment in time).41 In fact, the behavior
of 5 is characteristic of a two-state (one bit) memory or “flip-
flop” component in electronics.42 A flip-flop maintains its effect
on a system indefinitely until an input pulse operates on it,
causing its output to change to a new indefinitely stable state
according to defined rules.43 Different variants include T-, S-R-,
J-K-, and D-type flip-flops. We suggest that a molecule such
as 5 should be termed a “two-state” Brownian flip-flop be-
cause the substrate must exist in one of two compartments
(41) Combinational logic circuits (NAND, NOR, OR, XOR, EOR, two- and
three-input INH, etc.) are assembled by connecting combinations of AND,
NOT, and OR gates in various ways. These, in turn, are assembled by
connecting simple “on”-”off” switches in various ways. See: (a) Ben-
Ari, M. Mathematical Logic for Computer Science; Prentice-Hall: Hemel
Hempstead, 1993. For reviews on molecular-scale combinational logic
systems see: (b) Raymo, F. M. AdV. Mater. 2002, 14, 401-414 and (c) de
Silva, A. P.; McClenaghan, N. D. Chem. Eur. J. 2004, 10, 574-586. For
the sequential operation of combinational logic gates see: (d) de Silva, A.
P.; Dixon, I. M.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Maxwell, P. R.
S.; Rice, T. E. J. Am. Chem. Soc. 1999, 121, 1393-1394 and (e) Raymo,
F. M.; Giordani, S. Org. Lett. 2001, 3, 3475-3478.
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4069

A R T I C L E S
Chatterjee et al.
Scheme 9. Previously Reported8d Synthetic Molecular Rotary Motor 7 that Operates through a Full Brownian Ratchet Mechanism
in each molecule. However, there is an important difference
that arises because of the statistical nature of a two-state
Brownian flip-flop compared to the digital nature of its
electronic counterpart. If we wish to utilize the effect of many
molecules of 5 on a system, the operation of the flip-flop can
vary the bulk population distribution of the substrate between
the two compartments over a continuum from 85:15 to 44:56
fum:succ by modifying the balance-breaking reaction parameters
(e.g., by changing the time for which the photoisomerization
stimulus is applied). If, on the other hand, we use a single
molecule of 5 to influence a system, its effect is strictly
binary;the molecule is in one state or the other;with the fum:
succ ratios corresponding to the probability that the substrate
will be in a particular compartment given the history of the
inputs. In contrast, an electronic flip-flop is always utilized as
a single entity in a circuit and its effect is therefore always
binary.
machine to perform it.1 However, as function and mechanism
replace imagery as the driving force behind advances in this
field, the use of language needs to become more phenomeno-
logically based. For this reason, on the basis of the systems
described in this paper and those developed previously by
ourselves, Kelly, Feringa, Stoddart and others,8,10,13-18 we
suggest below definitions for four significant phenomenological
terms (ratcheting, escapement, balance, and linkage) that are
crucial for machines that operate by controlled Brownian motion.
Ratcheting is an often used, but previously ill-defined, process
in chemical terms. Unfortunately, this vagueness has led to the
term sometimes being applied to describe phenomena that are
unrelated to Brownian ratchet mechanisms. Escapement is the
counterpart to ratcheting and, as far as we are aware, has only
rarely been used to describe molecular-level events. The
statistical balance of a dynamic substrate and whether the parts
of the machine acting on the substrate allow exchange of the
substrate or not appear to be key factors that determine whether
the machine can perform a task or not. In fact, it appears that
the behavior of a molecular machine toward a substrate can be
defined by the changing relationship (linked/unlinked; balanced/
unbalanced) between the parts of machine interacting with the
substrate.
Language Necessary To Describe the Operation and
Mechanisms of Molecular-Level Mechanical Machines
Up to now the categorization of molecules as machines by
chemists has largely been iconic;the structures “look” like
pieces of machinery;or they are so-called because they carried
out a function that in the macroscopic world would require a
(i) “Ratcheting” is the capturing of a positional displacement
of a substrate through the imposition of a kinetic energy barrier
which prevents the displacement being reversed when the
thermodynamic driving force is removed. The key feature of
ratcheting is that the ratcheted part of the system is not linked
with (i.e., not allowed to exchange the substrate with) any part
of the system that it is ratcheted from. Ratcheting is a crucial
requirement for allowing a Brownian machine to be reset
without undoing the task it has performed. An example of
(42) (a) Malvino, A. P.; Brown, J. A. Digital Computer Electronics, 3rd ed.;
Glencoe: Lake Forest, 1993. (b) Mitchell, R. J. Microprocessor Systems:
An introduction; Macmillan: London, 1995.
(43) A series of stimuli-responsive molecular shuttles have been shown to exhibit
relatively long-lived nonequilibrium (“metastable”) states when operating
in self-assembled monolayers or a polymer electrolyte or at low temper-
atures (see refs 15i, 15j, and 15k). This may account for the junction
hysteresis observed in solid-state electronic devices that utilize such
rotaxanes (refs 20k-n). Kinetically stable nonequilibrium co-conformations
have also been observed in cyclodextrin-based shuttles; see refs 13l and
16d.
9
4070 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
Figure 5. Schematic representations of some already realized (a-c(i)) and possible (c(ii), d) types of multicompartment molecular-level machines. (a) A
two-state Brownian switch. (b) A two-state Brownian memory or flip-flop (shown operating through a partial energy ratchet mechanism; other mechanisms
are also possible to achieve the same machine function). (c) Examples of Brownian motors: (i) a two-stroke rotary motor; (ii) a three-compartment translational
motor. (d) Compartmentalized molecular-level machines that combine ratcheting and escapement with Boolean logic operations: (i) a four-compartment
Brownian machine that pumps a substrate in a given (variable) direction. Deciding which one of the red or green gates is used for ratcheting determines
whether the substrate is transported to the top compartment or the bottom. (ii) A four-compartment Brownian machine that sorts and separates different ions
(e.g., red ) Na+; blue ) K+). A logic operation providing selective access through each of the red and green gates ensures one type of ion is pumped into
each compartment.
ratcheting is the unlinking step 5D f 5E in Scheme 8. Because
it is used to kinetically stabilize an ultimately thermodynamically
unfavorable state, ratcheting is intrinsically associated with a
sequential logic sequence applied to a Brownian substrate.
(ii) “Escapement” is the (directional) release of a ratcheted
substrate in a statistically unbalanced system by lowering a
kinetic energy barrier (i.e., by linking). The key feature of
escapement is that it requires the linking of two unbalanced
parts of a system that were previously unlinked. An escapement
step must be subsequently ratcheted in order for a machine to
be able to do work repetitively on a substrate. An example of
escapement is the de-silylation step during the operation of the
irreversible molecular shuttle succ1-3 (Scheme 6, step (i)) or
the analogous step for system 5 (5B f 5C, Scheme 8).
(iii) “Balance” is the thermodynamically preferred distribution
of an exchangeable quantity or substrate over a machine or parts
of a machine. The impetus for net transportation of a substrate
comes from balance being broken. (Note that “balance” being
broken is not the same as “detailed balance” being broken;
“balance” is the thermodynamic driving force for “detailed
balance” to be broken.)
reactions rather than introducing or removing physical steric
or electronic barriers.
Types of Compartmentalized Molecular Machines
During development of the terminology necessary to describe
molecular-level behavior scientifically, the standard dictionary
definitions meant for everyday use are not always appropriate
for regimes that the definitions were never intended to cover.
As we have seen through the examples given in this paper, the
difference between a molecular motor and a molecular switch
is fundamental and profound because “motor” and “switch”
become different phenomenological descriptors at Brownian
length scales, not just iconic classifications of macroscopic
objects. As another example, simple switches that operate
through biasing Brownian motion do not stay in the same state
if the thermodynamic driving force is turned off and it is
important to distinguish them from other Brownian machines
that do. Indeed, we can identify three different fundamental types
of simple Brownian machines that act through various combina-
tions of balance-breaking, linking/unlinking, ratcheting, and
escapement steps;a Brownian switch (Figure 5a), a Brownian
flip-flop (Figure 5b), and a Brownian motor (Figure 5c). Each
of these machine types can also function as components in more
complex molecular-level machines (Figure 5d).
(iv) “Linkage” is the communication necessary for transporta-
tion of a substrate to occur between parts of a machine.
However, the ability to exchange the substrate between the
linked parts is not in itself enough for a task to be performed;
there must also be a driving force for it to occur (vide supra).
Linking and unlinking operations are purely kinetic parameters
and so can be accomplished by simply changing the rate of
(a) A “Two-state (or multistate) Brownian switch” is a
machine which can reversibly change the distribution of a
Brownian substrate (a moiety which undergoes Brownian
motion) between two (or more) distinguishable sites as a
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4071

A R T I C L E S
Chatterjee et al.
function of state of the machine. It does this by biasing the
Brownian motion of the substrate (Figure 5a). Classic stimuli-
responsive molecular shuttles, such as E/Z-1 and those listed
in refs 13-18 and 20, are examples of two-state (or three-state14)
Brownian switches.
heart of virtually every biological process. When we learn how
to build synthetic structures that can rectify random dynamic
processes and interface their effects directly with other molec-
ular-level substructures and the outside world, it will add a new
dimension to functional molecule and materials design. An
improved understanding of biological pumps, motors, and other
cellular machines will surely also follow.
(b) A “Two-state (or multi-state) Brownian flip-flop” is a
machine which can reversibly change the distribution of a
Brownian substrate between two (or more) distinguishable sites
and can be reset without restoring the original distribution of
the substrate (Figure 5b). The statistical distribution of the
substrate cannot be determined from the state of the flip-flop
(unlike a switch, which influences a substrate as a function of
state) but rather is determined by the history of operation of
the machine, i.e., through a form of sequential logic. Rotaxane
5 is an example of a two-state Brownian flip-flop. It operates
through a partial Brownian ratchet mechanism44 consisting of
the following steps applied to a pair of compartments: balance-
breaking, escapement, ratcheting, and resetting of the machine
(a second balance-breaking step). The original substrate distri-
bution is restored by an escapement-unlinking sequence.
(c) A “Brownian motor” is a machine which can repetitively
and progressively change the distribution of a Brownian
substrate, during which the machine is reset without restoring
the original distribution of the substrate (Figure 5c). Like a flip-
flop, a Brownian motor affects a system as a function of the
pathway that the machine takes, not as a function of state.
Catenane 7 (Scheme 9) is an example of a two-stroke rotary
Brownian motor (Figure 5c (i)).8d The substrate is repetitively
transported between two sites via two alternating pathways. We
shall shortly report on the synthesis and operation of a linear
three compartment Brownian motor (Figure 5c (ii)).45
Experimental Procedures for the Operation of
Machine-Substrate Systems 3 and 5
Machine-Substrate System 3. Step (i). A solution of [2]rotaxane
succ1-3 (80 mg, 0.057 mmol) in aqueous acetic acid (80%, 10 mL)
was heated at 60 °C for 1 h. The reaction mixture was concentrated
under reduced pressure and traces of acetic acid were removed with a
toluene azeotrope (3 × 50 mL). To remove non-rotaxane impurities,
the crude reaction mixture was filtered through a plug of silica gel
(neat dichloromethane followed by dichloromethane:methanol 95:5) and
the solvent removed under reduced pressure to furnish fully de-silylated
material.
Step (ii). The de-silylated product from step (i) was dissolved in
dichloromethane (20 mL) and treated with imidazole (70 mg, 1.0 mmol),
tert-butyl-dimethylsilyl chloride (90 mg, 0.60 mmol), and a catalytic
amount (5.0 mg) of 4-(dimethylamino)pyridine and the mixture stirred
at room temperature for 1 h. To remove non-rotaxane impurities, the
crude reaction mixture was filtered through a plug of silica gel (neat
dichloromethane followed by dichloromethane:methanol 98:2) and the
1
solvent removed under reduced pressure. The H NMR spectrum in
CDCl3 of the isomeric mixture of succ1-3 and succ2-3 resulting from
this two-step operation is shown in Figure 2c.
Machine-Substrate System 5. Step (i). A solution of [2]rotaxane
fum-E-5 (30 mg, 0.021 mmol) in aqueous acetic acid (80%, 2.0 mL)
was heated at 60 °C for 1 h. The reaction mixture was concentrated
under reduced pressure and traces of acetic acid were removed with a
toluene azeotrope (3 × 5 mL). To remove non-rotaxane impurities,
the crude reaction mixture was filtered through a plug of silica gel
(neat dicholormethane followed by dichloromethane:methanol 95:5) and
the solvent removed under reduced pressure to furnish fully de-silylated
material.
Step (ii). A solution of the de-silylated product from step (i) (20
mg) in dichloromethane (50 mL) was placed in a quartz vessel, degassed
with argon (3 × 5 min), and irradiated at 312 nm using a multilamp
photoreactor for five successive 5 min sessions. The progress of the
reaction was monitored by 1H NMR spectroscopy. The photostationary
state was reached after 20 min (4 × 5 min irradiations). The reaction
mixture was concentrated under reduced pressure and used directly in
the next step.
Step (iii). The de-silylated photoisomerized material (20 mg) was
dissolved in dichloromethane (2.0 mL) and treated with imidazole (20
mg, 0.29 mmol), tert-butyldimethylsilyl chloride (26 mg, 0.17 mmol),
and a catalytic amount (5.0 mg) of 4-(dimethylamino)pyridine. The
reaction mixture was stirred for 2 h, during which time a colorless
precipitate appeared. Water (2 mL) was added and the reaction mixture
extracted with dichloromethane (5 mL). The organic layer was
separated, washed with brine (saturated, 5 mL), dried over Na2SO4,
and concentrated under reduced pressure to furnish the crude re-silylated
isomeric mixture of rotaxanes. To remove non-rotaxane impurities, the
mixture was filtered through a plug of silica gel (petroleum ether:ethyl
acetate 8:2 followed by dichloromethane:methanol 8:2). This mixture
was concentrated under reduced pressure and used directly in the next
step.
(d) By combining the three fundamental Brownian machine
types which work through combinations of balance-breaking,
linking/unlinking, ratcheting, and escapement with other com-
binational and/or sequential operations based on Boolean logic,
machines that can carry out more complex functions, such as
variable, directional, travel (Figure 5d (i)), and “sorting and
separating” (Figure 5d (ii)), can be envisaged.
Apart from allowing a large amplitude one-dimensional
motion to be considered independently of other movements,
there is nothing special about rotaxanes in terms of mechanical
mechanisms.12,46 The relationships described in this paper are
applicable to any type of molecular-level construct, with
controlled motion possible in a kinetically associated structure
in two or three dimensions as well as just one. They may also
be useful in understanding the changes involved in biological
machines as they bring about movement, function, and the
transport of Brownian substrates (molecules and ions) in 1-D
channels, across membranes, and between parts of a protein.
To appreciate the technological potential of controlled mo-
lecular-level motion, one only has to consider that it lies at the
(44) Since the method of operation of 5 involves varying the relative heights of
energy minima and maxima irrespective of the position of the macrocycle,
this is a partial “energy ratchet” mechanism (ref 25b). Other types of
mechanisms are known (see ref 22), for example, an “information ratchet”
which achieves directional transport of a Brownian substrate by varying
the relative heights of energy maxima using information provided by the
position of the substrate (ref 25b).
(45) Chatterjee, M. N.; Goldup, S.; Kay, E. R.; Leigh, D. A. Manuscript in
preparation.
(46) Indeed, their branched topologies rule out rotaxanes as architectures for
the particular examples of compartmentalized machines shown in Figure
5d (i) and (ii).
Step (iv). To a solution of the re-silylated isomeric mixture of four
rotaxanes from step (iii) (20 mg) in dichloromethane (1.0 mL) was
added piperidine (100 µL, 1.0 mmol). The reaction was stirred for 12
h. To remove non-rotaxane impurities, the crude reaction mixture was
filtered through a plug of silica gel (neat dichloromethane followed by
9
4072 J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006

Compartmentalized Molecular Machines
A R T I C L E S
dichloromethane:methanol 8:2) and the solvent removed under reduced
pressure. The 1H NMR spectrum in CDCl3 of the isomeric mixture of
fum-E-5 and succ-E-5 obtained after the four-step operation is shown
in Figure 3d. An essentially identical 1H NMR spectrum was obtained
upon reversing the sequence of steps (i) and (ii).
Research Fellow and holds a Royal Society-Wolfson Research
Merit Award.
Supporting Information Available: Experimental procedures
and spectroscopic data for succ1-3 and fum-E-5 and their
precursors. This material is available free of charge via the
Internet at http://pubs.acs.org. See any current masthead page
for ordering information and Web access instructions.
Acknowledgment. This work was supported by the European
Union Future and Emerging Technology program Hy3M, the
EPSRC, and the Carnegie Trust. D.A.L. is an EPSRC Senior
JA057664Z
9
J. AM. CHEM. SOC. VOL. 128, NO. 12, 2006 4073