sign in sign up
<<
Home , Fraser Stoddart

PROFILE PROFILE ProfileofJean-PierreSauvage,SirJ.FraserStoddart, and Bernard L. Feringa, 2016 Nobel Laureates in Chemistry Jonathan C. Barnesa,1 and Chad A. Mirkinb,c,d,1

Mankind has long been fascinated by the discovery many to be the first electric motor. At that time, it was and invention of new technologies and gadgetry. not clear what application Henry’s invention might ful- Often times this fascination is borne out of necessity, fill, and even he admits in his manuscript to the editor but in many other cases, the sometimes lifelong pursuit that, “...the article, in its present state, can only be is that of pure intellectual curiosity. The beginning of considered a philosophical toy...” (4). Little did Henry the 19th century revealed as much when Alessandro know at the time that his discovery would later lead to Volta reported (1) the first electrical battery—now re- the development of more advanced electrical motors ferred to as a Voltaic cell—using stacks of copper and that are capable of doing work at then unfathomable zinc plates separated by a sulfuric acid electrolyte. This scales, and ultimately would result many years later in discovery of a constant electrical energy source later the development of complex machinery in the form of led to Hans Christian Oersted’s observation of electro- assembly lines, elevators, and modern-day all-electric magnetic fields in 1820 (2), Michael Faraday’s vertical motor vehicles, to name only a few. conducting wire orbiting around a magnet in 1821 (3), The 2016 Nobel Laureates in Chemistry (Nobelprize. and eventually Joseph Henry’sreport(4),in1831,ofa org) (Fig. 1)—namely, Jean-Pierre Sauvage, Sir J. Fraser small rocker actuated by reciprocating magnetic attrac- Stoddart, and Bernard L. Feringa—may be likened to tion and repulsion, the latter of which is considered by that of Volta, Oersted, Faraday, and Henry in terms of how they initiated completely new areas of research, for which they have been honored with this year’sNobel Prize, specifically for their seminal contributions to “the design and synthesis of molecular machines” (5). These researchers’ fundamental explorations into the realms of mechanically bonded molecules and sterically crowded olefins changed the way think about molecular systems and paved the way for the possibility to design more complex artificial molecular machinery. Their work over the past three decades is, in part, responsible for putting chemists—synthetic organic ones in particular— at the forefront of the burgeoning fields of nanoscience and . At the very least, these men intro- duced a fundamentally new way of constructing molec- ular matter at the nanoscale, and at the very most they laid the groundwork for a new field focused on using such capabilities to realize complex interlocked mole- cules with extraordinary functions we typically equate with macroscopic machinery. In this profile we highlight the seminal achievements in Chemistry of each of the 2016 Nobel Laureates, and we discuss how their research has evolved and impacted greatly the field Fig. 1. 2016 Nobel Laureates in Chemistry shortly after arriving in Stockholm: over the past 30 years. (Left to Right) Bernard L. Feringa, Sir J. Fraser Stoddart, and Jean-Pierre The story of molecular machines begins with Jean- Sauvage. Image courtesy of Stuart Cantrill (photographer). Pierre Sauvage, who in 1971 obtained his doctorate

aDepartment of Chemistry, Washington University, St. Louis, MO 63130; bDepartment of Materials Science and Engineering, , Evanston, IL 60208; cInternational Institute of Nanotechnology, Northwestern University, Evanston, IL 60208; and dDepartment of Chemistry, Northwestern University, Evanston, IL 60208 This article is part of a series of articles in PNAS highlighting the discoveries and profiling the award winners of the . Articles about the 2016 Nobel Laureates in Physiology or Medicine, and in Physics can be found on page 201 in issue 2 of volume 114 and on page 626, respectively. Author contributions: J.C.B. and C.A.M. wrote the paper. The authors declare no conflict of interest. 1To whom correspondence may be addressed. Email: [email protected] or [email protected].

620–625 | PNAS | January 24, 2017 | vol. 114 | no. 4 www.pnas.org/cgi/doi/10.1073/pnas.1619330114 Downloaded by guest on September 24, 2021 from the University of Strasbourg while developing (6) foundation for Stoddart’s seminal report (15) in the first cryptand-based ion receptors under the super- 1989 of the first template-directed synthesis that used vision of Jean-Marie Lehn [corecipient of the 1987 through-space donor–acceptor interactions (Fig. 2B) , with Donald Cram and to synthesize an all-organic [2] in 70% overall Charles Pederson, “for their development and use of yield. Analogous to Sauvage’s “threading” approach, molecules with structure-specific interactions of high Stoddart’s group also used (Fig. 2B)a“closed” crown selectivity” (7)]. Before Sauvage’s doctoral studies, ether macrocycle 6 that was added to an existing mix- + the synthesis of interlocked structures—specifically [2] ture of the bis(pyridinium) salt 72 and 1,4-bis(bromo- , which consist of two noncovalently inter- methyl)benzene (8). After one of the free pyridine rings + locking molecular rings—had been achieved either of 72 had undergone a nucleophilic substitution reac- + through a low-yielding statistical protocol (8), or by tion with 8 to afford 73 , the tricationic donor–acceptor + using a directed covalent approach (9) that was quite complex 73 ⊂6 was formed, which immediately pro- laborious, because it required many synthetic steps ceeded to the ring-closing step in the one-pot reac- and afforded only modest overall product yields. In tion and generated cyclobis(paraquat-p-phenylene) + + 1983, while a professor at the University of Strasbourg, (CBPQT4 , 94 ) in the process, thus producing the + Sauvage demonstrated (Fig. 2A) the first high-yielding desired donor–acceptor [2]catenane 104 . This effi- template-directed synthesis of a [2]catenane by capital- cient synthetic protocol made it possible for the first izing on a “threading” approach where the formation of time to contemplate synthesizing MIMs on a large + a tetracoordinate complex 2⊂1 is achieved between a scale and to do so by using a combination of noncova- “closed” phenanthroline-based macrocycle 1 and an lent bonding interactions (specifically, C–H···O, πD···πA, “open” phenanthroline diol 2 centered about a cop- and C–H···π) that would later come into play in develop- per(I) metal atom (10). The stability of this complex is ing molecular switches/machines in the form of sufficiently high to allow for the ring-closing step, which bistable MIMs. proceeds through two nucleophilic substitutions of Following up on their seminal reports of the a diiodo oligoethylene glycol 3, resulting in a 42% template-directed syntheses of the pair of [2]catenanes + + yield for the desired [2]catenate 4 .Sauvage’sgroup 5 and 104 , the two researchers subsequently devel- demonstrated (11) later that the copper(I) atom could oped a wide range of MIMs over the next several years be removed either chemically with KCN, or through (16). Of particular importance to the field of molecular electrochemical oxidation of copper(I) to copper(II). machines is the independent work published sepa- The loss of the copper atom from the center of the [2] rately in 1994 by both research groups. Specifically, catenate affords a [2]catenane 5, where each phenan- Stoddart’s group reported (17) a key finding in the throline subunit rotated away from one another. Al- journal , in which they described the synthesis though this first example of using a template-directed and operation of the first bistable, donor–acceptor + protocol produced what many would consider to be a [2] (114 )(Fig.2C). Molecular motion within + + molecular “switch” rather than a “machine,” the impor- 114 was demonstrated by toggling the CBPQT4 tance of Sauvage’s seminal work cannot be overstated. ring back and forth in a controlled fashion along By providing a practical blueprint on how to synthesize the linear dumbbell portion of the molecule stopping mechanical bonds—a new form of bonding in chemistry at the benzidene and biphenol recognition units, that combines two different molecules into one— where the benzidene moiety is more π-electron–rich Sauvage’s group uncovered the path that would ul- and thus a more natural starting position (i.e., 84:16 timately lead others to the development of the field ratio of benzidene:biphenol) for the π-electron–deficient + of mechanically interlocked molecules (MIMs) and CBPQT4 ring. Control over the translational motion of ultimately to the discovery of molecular machines. the tetracationic ring was achieved either through elec- Several years before Sauvage’s critical discovery, trochemical oxidation of benzidene, or through the ad- + J. Fraser Stoddart had obtained his doctorate in 1966 dition of acid, which when protonated (116 )(Fig.2C) from University and, after a stint at Queen’s created enough Coulombic repulsion to force the + University in Canada as a National Research Council CBPQT4 ring onto the biphenol station. The process Postdoctoral Fellow, he began his independent career was shown to be reversible simply by adding an ex- in 1970 at Sheffield University as an Imperial Chemical cess of pyridine, and thus the first rotaxane-based Industries Research Fellow before joining the faculty was born. Also in 1994, Sauvage’s there as a Lecturer in Chemistry. Early on in his ca- group published a communication in the Journal of reer, Stoddart’s research focused on the chemistry of the American Chemical Society (JACS), where they de- carbohydrates (12), and then it transitioned in 1974 scribed the electrochemically induced switching of a + into the synthesis of macrobicyclic polyethers that copper-based [2]catenate (124 )(Fig.2D) (18). The mo- + possessed carbon bridgeheads (13), analogous to lecular composition of 124 (where the subscripted the nitrogen-based compounds reported earlier value 4 refers to the coordination number of the copper by Dietrich et al. (6). Stoddart then proceeded to center) is similar to that of 5; however, an additional publish many papers that described the binding of metal binding site is introduced in the form of a transition metal ammines by various crown ether deriv- 2,2′,6′,2′′-terpyridine (terpy). The addition of the triden- atives (14). These studies ultimately led Stoddart’sgroup tate terpy ligand is important in terms of the switching + 2+ to investigate the binding of π-electron–deficient mechanism, whereby upon oxidation of 124 to 124 paraquat (also known as methyl viologen) and diquat at the applied electrical potential of +0.67 V, copper(I) is guests using bisparaphenylene[34]crown-10 (6 in oxidized to copper(II), and the change in the oxidation Fig. 2B) as the host. This latter discovery laid the state of the metal center resulted in the rotation of the

Barnes and Mirkin PNAS | January 24, 2017 | vol. 114 | no. 4 | 621 Downloaded by guest on September 24, 2021 Fig. 2. (A) The first template-directed synthesis of a copper-complexed [2]catenane using metal coordination by Sauvage and coworkers, followed by removal of copper(I) after oxidizing it electrochemically to copper(II). (B) Stoddart’s donor– acceptor template-directed one-pot synthesis of a [2]catenane. (C) The reversible shuttling between “stations” of an electron-deficient cyclophane along a dumbbell-shaped molecular track was published by Stoddart and coworkers in 1994 (17). (D) Also in 1994, Sauvage’s group demonstrated the concept of a “swinging catenane” by introducing an additional metal binding site (18). Reproduced from ref. 11, with permission of The Royal Society of Chemistry. (E)An artificial molecular pump acts against equilibrium to store cyclophane molecules along a connected aliphatic chain attached to the pump. Reproduced from ref. 19, with permission from Macmillan Publishers Ltd.: Nature Nanotechnology, copyright 2015. (F) Sauvage’s artificial molecular muscle was based on extension and contraction of a double rotaxane architecture in the presence of monovalent or divalent transition metals, respectively. Reproduced from ref. 11, with permission of The Royal Society of Chemistry. DMF, dimethylformamide; RT, room temperature.

terpy-containing ring (albeit slowly) to its five-coordinate MIM, the demonstration of this type of electrochemically 2+ conformation in 125 . The rotation process was further induced motion represents the first classic example of examined by Sauvage and coworkers by electrochemi- a molecular switch based on a [2]catenate architecture. cally reducing the copper(II) back to copper(I) at an ap- Significant advances in the area of MIMs and + plied potential of –0.06 V, thus generating 125 ,which molecular switches/machines have been made by + rearranges quickly back to the more favorable 124 .Al- both research groups since each of the aforementioned though the kinetics were a bit slow for this copper-based investigations were first reported. Two such examples

622 | www.pnas.org/cgi/doi/10.1073/pnas.1619330114 Barnes and Mirkin Downloaded by guest on September 24, 2021 are illustrated in Fig. 2 E and F, where Stoddart’sfocus the initial proof-of-concept that a thermally stable, in more recent years has revolved around developing yet light-activated chiroptical molecular switch nonequilibrium molecular pumps (19) (Fig. 2E) that cap- was possible. italize on favorable radical–radical pairing interactions Over the years the design for Feringa’s chiroptical + between chemically reduced CBPQT4 and viologen molecular switch was improved (23), and in 1999 a subunits to collect and store the tetracationic rings specific design was implemented that was similar to along an attached aliphatic chain. Sauvage’s interest the very first sterically overcrowded alkene reported over the years grew toward systems like that shown in by Feringa in 1977, except with one major modification: Fig. 2F, whereby metal-based rotaxane dimers (20), instead of having hydrogen atoms in the 4,4′ positions possessing both phenanthroline and terpyridine metal of the phenanthrylidene groups, they replaced them + binding sites, allowed for the extension (132 ) and con- with methyl groups [Fig. 3 A and C;see(P,P)-trans-16, + traction (134 ) of the double MIM in response to for example]. The stereogenic centers (R for both whether monovalent or divalent metals were present, halves) bearing the methyl groups, and the methyl’s respectively. This type of artificial corresponding axial/equatorial orientations, were criti- can be likened to the motion that takes place between cally important for the first demonstration of unidirec- actin and myosin filaments during the actuation of hu- tional motion of a molecular motor. Starting from the man muscles, and therefore represents a significant ad- pure (P,P)-trans-16 isomer (Fig. 3C), the first stage of vance in terms of designing future molecular machines. the rotation was initiated by irradiating it at or above The previous examples of translational or rotational 280 nanometers while at –55 °C to stop the rotation motion of molecular rings in rotaxane- and catenane- at the next stage. This photo-induced isomerization based molecular systems by Stoddart and Sauvage of the double bond resulted in a trans-to-cis confor- certainly jump-started the field and this is primarily mational change that placed the methyl substituents because of the advent of the mechanical bond. If in an equatorial position and a population distribu- practical molecular machines are ever to become a tion of 95% (M,M)-cis-16 to only 5% of (P,P)-trans-16,a reality, however, a wider range of controllable molecular considerable improvement over their first reported chi- motions are needed. This is where Bernard Feringa’s roptical switch in 1991. Allowing (M,M)-cis-16 to warm rotary molecular motors became important to the to 20 °C, initiated an irreversible helical inversion to field, and it is largely because of the ability of his the more energetically favored (P,P)-cis-16 conformer, later molecular rotor designs to undergo light-driven favored by the fact that the methyl substituents adop- unidirectional motion. ted an axial orientation once again. From (P,P)-cis-16, The key design feature that is common to all of irradiation again isomerizes the double bond, which Feringa’s molecular motors is that of a sterically over- generates a 90% (M,M)-trans-16 to 10% (P,P)-cis-16 crowded alkene (21) (see Fig. 3A, cis-14 and trans-14), photostationary state, where the methyls are now equa- the synthesis for which was first developed in 1977 torial. In the last step of the rotation mechanism, the when Feringa was a graduate student under the super- 90:10 mixture is heated to 60 °C, which provides the vision of Hans Wynberg at the University of Groningen. rotor with enough energy to overcome the barrier for The advantage of incorporating an overcrowded alkene helical inversion from (M,M)-trans-16 to the lowest en- into a molecular motor is based on the fact that the ergy conformer (P,P)-trans-16. When irradiation of the energy barrier to interconvert between helical isomers propeller-like motor and the temperature (at 60 °C) are (i.e., P and M) is too great to be overcome at room held constant, the rotor experiences completely unin- temperature alone. This lack of racemization meant it terrupted unidirectional circular motion about the ste- was possible to isolate exclusively each enantiomeric rically crowded olefin. This remarkable study illustrated form by HPLC, followed by the step-wise investigation that through careful design, it was indeed possible to into the externally triggered rotary motions of the mol- establish a potentially useful nano-sized rotor that could ecule. After obtaining his doctorate in 1978, Feringa be controlled externally using light and heat. went to work for Shell Laboratories in Amsterdam, Even more remarkable is that in 2002, Feringa’s and later in Sittingbourne in the , be- group demonstrated that the biphenanthrylidene- fore becoming a Lecturer in at the based molecular motor (P,P)-trans-16 (Fig. 3A) could University of Groningen in 1984. In 1988, Feringa was be added at ∼2–6 wt% in a liquid-crystal (LC) matrix appointed successor of his doctorate advisor, Hans and still undergo unidirectional rotary motion that in- Wynberg, and it was around this time in 1991 that the duced color changes in the reorganized LC material first chiroptical molecular switch (22) was born (Fig. 3A, (24). Improving on this feat, Feringa’s group demon- under 1991). The design of this switch involved com- strated in 2006 that a newly designed molecular motor bining a methoxythioxanthylidene on one side of the 17 (Fig. 3A, under 2006) added (at 1 wt%) to an LC film overcrowded alkene and a bulky tetrahydrophenan- was also capable of causing rotational reorganization threne on the other side. Irradiation of pure (M)-cis-15 of the LC material (25). On top of that (literally!), they with light at 300 nanometers resulted in an isomeriza- showcased (Fig. 3A, Inset) the light-driven molecular tion (Fig. 3B)to(P)-trans-15 at a ratio of 64% (M)-cis-15 motor’s ability to rotate large objects—a glass rod and 36% (P)-trans-15, which is referred to as a photo- 10,000 times the size of the motor in this case—that stationary state in the JACS communication authored was placed on top of the LC film. This demonstration is by Feringa and coworkers (22). Irradiation at a higher incredible when one thinks about how the molecular- energy (i.e., 250 nanometers) resulted in a slightly dif- scale rotations about the double bond of the motor ferent (M)-cis-15 to (P)-trans-15 ratio of 68–32%, re- (present only up to 1 wt%) were translated so effec- spectively. This first example of a molecular rotor was tively to the macroscopic level.

Barnes and Mirkin PNAS | January 24, 2017 | vol. 114 | no. 4 | 623 Downloaded by guest on September 24, 2021 Fig. 3. (A) The birth and evolution of some of Feringa’s key molecular motors. The Inset under the 2006 rotor is an experiment where Feringa’s group added 1 wt% of 17 to an LC film and demonstrated rotation of a glass rod 10,000 times the size of the rotor. (Inset) Reproduced from ref. 25, with permission from Macmillan Publishers Ltd.: Nature, copyright 2006. (B) The isomerizations and helical inversions associated with the first chiroptical molecular switch. (C) Going back to the original 1977 rotor design, except for the addition of two methyl groups in the 4,4′ positions, Feringa’s group demonstrated light-driven unidirectional rotation of a molecular rotor for the first time. (D) Incorporating a modified 2006 rotor into a diyne-bridged molecule, Feringa and coworkers developed the first externally triggered displacement of a “nanocar” across a Cu(111) surface. Reproduced from ref. 26, with permission from Macmillan Publishers Ltd.: Nature, copyright 2011.

Another beautiful example of a Feringa molecular of chemistry, and particularly the field of molecular motor in action came in 2011 when his group—after machines, a field which they initiated and have been carrying out a multistep synthesis and isolation of the torch-bearers of for so many years. At the heart of the resultant isomers—incorporated four molecular their individual accomplishments and discoveries, motors at the corners of a diyne-bridged molecule however, lies the pure intellectual joy associated with (Fig. 3D) that resembled the shape (or at least the the making and measuring of organic compounds. chassis) of a car (26). After subliming the four-wheeled In today’s current funding climate, it can some- molecule on a Cu(111) surface at 553 K, the re- times be difficult to pursue purely fundamental sci- searchers proceeded to activate the motors of the ence; however, Sauvage, Stoddart, and Feringa have meso-(R,S-R,S) isomer using a scanning tunneling shown us that it is not only okay to do so, but that this microscope tip (Fig. 3D, Top Right) that served the role type of pursuit can be incredibly enriching along of the light and thermal energy sources required for the way. The Royal Swedish Academy of Sciences isomerization/inversion (Fig. 3D, Bottom Right). After 10 excitations of this isomer, the researchers were able agreed, and validated their lifelong efforts by award- to observe relatively linear movement over a distance ing them the 2016 Nobel Prize in Chemistry. Who of 6 nanometers. This translates to many revolutions knows how far it will go, but hopefully these great — per motor to span that distance, which is by no means chemistry pioneers will have the opportunity as a small feat. Joseph Henry did with his invention of the first elec- Looking back on the work that was done over the tric motor—to witness the full extent of how their last three decades by this year’s trio of Chemistry early endeavors have impacted, and will continue Nobel Prize winners, it is quite remarkable to con- to impact, the many generations of scientists that template just how much they changed the landscape follow in their footsteps.

1 Trasatti S (1999) 1799–1999: Alessandro Volta’s ‘electric pile’—Two hundred years, but it doesn’t seem like it. J Electroanal Chem 460(1–2):1–4. 2 Stern DP (2002) A millennium of geomagnetism. Rev Geophys 40(3):1-1–1-30. 3 Gooding DC (1996) Scientific discovery as creative exploration: Faraday’s experiments. Creat Res J 9(2-3):189–205. 4 Henry J (1831) On a reciprocating motion produced by magnetic attraction and repulsion. Am J Sci 20(2):340–343. 5 Nobel Prize (2016) The Nobel Prize in Chemistry. Available at www.nobelprize.org/nobel_prizes/chemistry/laureates/2016/. Accessed November 20, 2016. 6 Dietrich B, Lehn JM, Sauvage JP (1969) Les cryptates. Tet Lett 34:2889–2892. 7 Nobel Prize (2016) The Nobel Prize in Chemistry. Available at www.nobelprize.org/nobel_prizes/chemistry/laureates/1987/. Accessed November 20, 2016. 8 Wasserman E (1960) The preparation of interlocking rings: A catenane. J Am Chem Soc 82(16):4433–4434.

624 | www.pnas.org/cgi/doi/10.1073/pnas.1619330114 Barnes and Mirkin Downloaded by guest on September 24, 2021 9 Schill G, Lüttringhaus A (1964) Preparation of catena compounds by directed synthesis. Angew Chem Int Ed Engl 3:546–547. 10 Dietrich-Buchecker CO, Sauvage JP, Kintzinger JP (1983) Une nouvelle famille de molecules: Les metallo-catenanes. Tetrahedron Lett 24(46):5095–5098. 11 Durot S, Reviriego F, Sauvage JP (2010) Copper-complexed catenanes and in motion: 15 years of molecular machines. Dalton Trans 39(44):10557–10570. 12 Stoddart JF, Jones JKN (1968) Some structural features of citrus limonia gum (lemon gum). Carbohydr Res 8:29–42. 13 Burden IJ, Stoddart JF (1974) Constitutional isomerism in bicyclic diacetals and the conformational behavior of cis-fused 1,3,6,8- tetraoxabicyclo[5.3.0]decanes. J Chem Soc Chem Commun (21):863–864. 14 Colquhoun HM, Stoddart JF (1981) Second-sphere co-ordination of neutral and cationic transition metal complexes by crown ethers. J Chem Soc Chem Commun (12):612–613. 15 Ashton PR, et al. (1989) A [2]catenane made to order. Angew Chem Int Ed Engl 28(10):1396–1399. 16 Bruns CJ, Stoddart JF (2016) The Nature of the Mechanical Bond: From Molecules to Machines (Wiley & Sons, Hoboken, NJ). 17 Bissell RA, C ´ordovaE, Kaifer AE, Stoddart JF (1994) A chemically and electrochemically switchable . Nature 369:133–137. 18 Livoreil A, Dietrich-Buchecker CO, Sauvage JP (1994) Electrochemically triggered swinging of a [2]-catenate. J Am Chem Soc 116(20): 9399–9400. 19 Cheng C, et al. (2015) An artificial molecular pump. Nat Nanotechnol 10(6):547–553. 20 Jimenez ´ MS, Dietrich-Buchecker CO, Sauvage JP (2000) Towards synthetic molecular machines: Contraction and stretching of a linear rotaxane dimer. Angew Chem Int Ed 39(18):3284–3287. 21 Feringa B, Wynberg H (1977) Torsionally distorted olefins. Resolution of cis- and trans-4,4′-Bi-1,1′2,2′3,3′- hexahydrophenanthrylidene. J Am Chem Soc 99(2):602–603. 22 Feringa BL, Jager WF, Lange B (1991) Chiroptical molecular switch. J Am Chem Soc 113(14):5468–5470. 23 Koumura N, Zijlstra RWJ, van Delden RA, Harada N, Feringa BL (1999) Light-driven monodirectional molecular rotor. Nature 401(6749):152–155. 24 van Delden RA, Koumura N, Harada N, Feringa BL (2002) Unidirectional rotary motion in a liquid crystalline environment: Color tuning by a molecular motor. Proc Natl Acad Sci USA 99(8):4945–4949. 25 Eelkema R, et al. (2006) Molecular machines: Nanomotor rotates microscale objects. Nature 440(7081):163. 26 Kudernac T, et al. (2011) Electrically driven directional motion of a four-wheeled molecule on a metal surface. Nature 479(7372): 208–211.

Barnes and Mirkin PNAS | January 24, 2017 | vol. 114 | no. 4 | 625 Downloaded by guest on September 24, 2021