Enantioselective room temperature phosphorescence detection of non-phosphorescent analytes based on interaction with beta-cyclodextrin/1-bromonaphthalene complexes

Talan a 66 (2005) 634–640 Enan ioselec i e oom empe a u e phospho escence de ec ion o non-phospho escen analy es based on in e ac ion wi h ␤-cyclodex in/1-b omonaph halene complexes Ca men Ga c´ ıa-Ruiz1, XueShan Hu2, F eek A iese, Cees Gooije ∗ Depa men o Analy ical Chemis y and Applied Spec oscopy, Lase Cen e, V ije Uni e si ei , De Boelelaan 1083, NL-1081 HV Ams e dam, The Ne he lands Recei ed 3 June 2004; ecei ed in e ised o m 30 Sep embe 2004; accep ed 1 Decembe 2004 A ailable online 7 Janua y 2005 Abs ac Men hol (MT) induces s ong oom empe a u e phospho escence (RTP) o 1-b omonaph halene (1B N) in aqueous ␤-cyclodex in (␤-CD) suspensions, e en unde non-deoxygena ed condi ions. In e es ingly, (−)-MT and (+)-MT enan iome s gi e ise o di e en phospho escence in ensi ies, he di e ence being 19±3%. I is a gued ha he signal can be mainly asc ibed o he o ma ion o e na y complexes ␤- CD/1B N/MT which show di e en RTP li e imes, i.e. 4.28±0.06 and 3.71±0.06ms o (−)-MT and (+)-MT, espec i ely. Mos p obably, he s e eochemical s uc u e o (−)-MT p o ides a be e p o ec ion o 1B N agains quenching by oxygen han (+)-MT. This in e p e a ion is in line wi h he obse a ion ha unde deoxygena ed condi ions he phospho escence in ensi y di e ence o he wo complexes becomes e y small, i.e. only abou 4%. The li e ime di e ence unde ae a ed condi ions enables he di ec de e mina ion o he MT s e eochemis y. Fo mix u es, in iew o he 0.06ms unce ain y in he li e ime, enan iome ic pu i y can be de e mined down o 10%. Fu he mo e, in he case o MT he concen a ion o he leas abundan enan iome should be a leas 3×10−4M, since o he wise complex dissocia ion would obscu e he li e ime di e ence. © 2004 Else ie B.V. All igh s ese ed. Keywo ds: Men hol; Te na y complex; Chi al disc imina ion; Phospho escence li e ime 1. In oduc ion The obse a ion o oom empe a u e phospho escence (RTP) in liquid solu ions o suspensions gene ally equi es ho ough deoxygena ion. In e es ingly, he e a e excep ions o his ule: in ense RTP o 1-b omonaph halene (1B N) in aqueous ␤-cyclodex in (␤-CD) solu ions has been ob- se ed unde ae a ed condi ions in he p esence o alco- hols ( e -bu yl alcohol, cyclohexanol, cyclopen anol, and 1-pen anol) [1–3], alkanes (cyclohexane, b omocyclohex- ∗Co esponding au ho . Tel.: +31 20 4447524; ax: +31 20 4447543. E-mail add ess: gooije @ ew. u.nl (C. Gooije ). 1P esen add ess: Dep . o Analy ical Chemis y, Uni e si y o Alcal´ a, Spain. 2P esen add ess: Beijing Mic o-Tech Ins i u e, China. ane, and 1,2-dib omoe hane) [3], su ac an s ( i on X-100, sodium dodecyl sul a e, dodecylbenzene sul ona e, and de- oxychola e) [2,4,5], o polyme s (p-oc ylpolye hylene gly- col phenyle he ) [6]. This phenomenon can be a ibu ed o he o ma ion o e na y complexes ha p o ec he phos- pho opho e (1B N) in he exci ed iple (T1) s a e om quenchingbyoxygen.Basedon he o ma iono such e na y complexes, also a ious b omonaph ol de i a i es ha e been de ec ed by RTP unde non-deoxygena ed condi ions [7,8]. Naza o e al.ha ee en epo eds ongsignals o ano ganic phospho opho e no subs i u ed wi h a hea y a om such as b omine: o naph halene-␤-CD-adaman ane suspensions a phospho escence li e ime as long as 10.3s could be ob ained [9]. Only e y ecen ly, some applica ions based on RTP in ae a ed aqueous suspensions ha e been de eloped. Zhai e 0039-9140/$ – see on ma e © 2004 Else ie B.V. All igh s ese ed. doi:10.1016/j. alan a.2004.12.009 C. Ga c´ıa-Ruiz e al. / Talan a 66 (2005) 634–640 635 Fig. 1. S uc u e o he enan iome s o MT. al. ha e demons a ed he use o 6-b omo-2-naph hol/␤-CD as an RTP senso o 28 small o ganic molecules, showing a high molecula ecogni ion abili y o cyclohexane [10]. Sanz-Medel and co-wo ke s de eloped he i s RTP senso o pHbasedon helong- angeF¨ o s e ene gy ans e oma phospho opho e (1B N o 6-b omo-2-naph yl sul a e, which emi s RTP in ae a ed aqueous solu ions) o a mix u e o pH- sensi i e dyes [11]. The Sanz-Medel g oup has also epo ed wodi e en analy icalme hods o hede e mina iono me - cu y [12] o cyanide [13] in wa e samples based on he RTP o 1B N/␤-CD. Chi al compounds ha show na i e phospho escence can be dis inguished by RTP i hey o m complexes wi h chi- al ligands and i he esul ing dias e eoisome ic species ha e di e en spec oscopic p ope ies. Fo ins ance, we ha e demons a ed he enan ioselec i e de ec ion o phos- pho escen analy es (e.g., campho quinone) in ␣-CD solu- ions, based on di e ences in phospho escence li e imes (see accompanying pape [14]). In he p esen pape we will con- side he po en ial o RTP emi ed by 1B N/␤-CD complexes o dis inguish be ween he enan iome s o a chi al, non- phospho escen compound. We selec ed o his s udy he enan iome s o men hol (MT; see Fig. 1 o s uc u es), a compound ha dly de ec able in he liquid phase by con en- ionalspec oscopicmeanssincei doesno con ainanych o- mopho ic g oup. 2. Expe imen al sec ion 2.1. Chemicals and samples (1R,2S,5R)-(−)-MT, (1S,2R,5S)-(+)-MT, 1-b omonaph- halene (1B N), and ␤-cyclodex in (␤-CD) we e pu chased om Ald ich (S einheim, Ge many). Wa e used o he p epa a ion o he solu ions was pu i ied h ough a Milli-Q sys em om Millipo e (Bed o d, MA, USA). A s ock solu ion wi h 1B N and ␤-CD a nominal con- cen a ions o 10−4and 10−2M, espec i ely, was p epa ed by weighing he app op ia e amoun s, adding Milli-Q wa e , and s i ing o 1h. ␤-CD would be soluble a ha le el, bu in hep esence o 1B N a cloudysuspensionwasob ained,in acco dance wi h he li e a u e [12,13]. Howe e , no p ecip- i a ion was obse ed, indica ing ha he o al concen a ion (dissol ed+suspended) was s ill equal o he nominal con- cen a ion. Aqueoussolu ionso MTenan iome s(10−3M)we ep e- pa ed by weighing he app op ia e amoun o solid, adding Milli-Q wa e , and sonica ing o 30min un il o al dissolu- ion. Samples we e p epa ed by mixing 1.5mL o he 1B N/␤- CDsuspensionwi h1.5mLo anaqueoussolu ioncon aining MTin hedesi edconcen a iono enan iome iccomposi ion and o exing o 30min. Du ing his p ocess he cloudy so- lu ionclea ed,indica ingsu icien solubili yo he inalmix- u e. In all expe imen s he inal concen a ion o 1B N was 5×10−5M and ha o ␤-CD was 5×10−3M; he concen- a ion o MT was 5×10−4M unless o he wise indica ed. We ound ha a e 2 weeks o s o age in he e ige a o (∼4◦C), phospho escence li e imes would inc ease, indica - ing g adual agg ega ion. In p ac ice, suspensions we e mea- su ed wi hin 1–4 days o p epa a ion, and he li e imes we e checked be o e each se ies o measu emen s. 2.2. Ins umen a ion An LS-50B luminescence spec ome e (Pe kin-Elme , Beacons ield, UK) equipped wi h a lash xenon a c sou ce wasused o phospho escencemeasu emen s.Exci a ion and emission sli s we e 10nm; exci a ion and emission wa e- leng hs we e se a 285 and 528nm, espec i ely. In o de o emo e in e e ing luo escence a delay ime o 0.10ms was used; he ga ing ime was ypically 5.00ms. Fo he li e- ime measu emen s he ga ing ime was se a 0.10ms, while he delay imes we e a ied in s eps o 0.10ms (1B N/␤- CD solu ions) o in s eps o 0.5ms (1B N/␤-CD solu ion in he p esence o men hol enan iome s). T iple li e ime alues we e ob ained by i ing he RTP decay cu es o a mono- o bi-exponen ial decay cu e (O igin 6.1, O iginLab Co po a- ion). Fo compa ison, phospho escence spec a unde deoxy- gena ed condi ions we e eco ded in a special long-neck lu- minescence cu e e (Hellma Benelux BV, The Ne he lands) wi h a cap and Te lon ubes, so ha a low ni ogen low h ough he suspension could be main ained. Fluo escence measu emen s we e pe o med using an exci a ion wa e- leng h o 276nm, an emission wa eleng h a 336nm, and exci a ion and emission sli s o 10nm. 3. Resul s and discussion I should be no ed ha 1B N/␤-CD complexes in aque- oussys emsa e p one o agg ega e o ma ion, aphenomenon ha ob iously will also a ec he e na y complex o ma ion wi h men hol. The e o e, ca e ul sample p epa a ion p oce- du es had o be de eloped in o de o achie e ep oducible esul s. I was obse ed ha mixing o he solu ions could 636 C. Ga c´ıa-Ruiz e al. / Talan a 66 (2005) 634–640 no be pe o med by sonica ion since he accompanying em- pe a u e ise a ec ed he esul s; o exing u ned ou o be he mixing me hod o choice. We in es iga ed he in luence o he o exing ime on he phospho escence in ensi y o 1B N/␤-CD in he p esence o men hol enan iome s. The in- ensi y inc eased du ing he i s 10min, hen le elled o and a mo e o less s able signal was ob ained o o exing imes up o 60min. Fo p ac ical easons, all suspensions we e o - exed o 30min be o e he spec a we e eco ded. Inlinewi h he li e a u e [1], upon addi ion o men hol we p opose he o ma ion o a e na y complex1B N/␤-CD/MT. In e es ingly, Fig. 2 ( ull lines) shows ha he phospho es- cence in ensi ies o he 1B N/␤-CD suspension in he p es- ence o he wo MT enan iome s a e di e en , he di e ence being 19±3%. In con as , he much weake luo escence spec a o he complex in he p esence o he wo MT enan- iome s we e o equal in ensi y (spec a no shown). These esul scanbeunde s oodi weassume ha hecomplexs uc- u es o 1B N/␤-CD/MT a e no iden ical o (−)- and (+)- MT, and ha hese di e ences a ec he phospho escence yields, bu no he luo escence yields. P esumably he e is a di e ence in p o ec ion agains oxygen quenching. To es hishypo hesis,li e imes o he aqueous 1B N/␤-CDsuspen- sion in absence and p esence o men hol enan iome s we e measu ed. As shown in Fig. 2 (lowe spec um), also in absence o men hol does 1B N/␤-CD show RTP in non-deoxygena ed aqueoussuspension, hougho a ela i elylowin ensi y.This indica es ha also o he bina y complex he e is al eady a subs an ial p o ec ion o he T1-s a e o 1B N, ully in line wi h he indings epo ed by Tu o e al. in he ea ly 1980’s [15]. The associa ed phospho escence decay cu e is shown in Fig. 3. I clea ly i s o a bi-exponen ial decay as demon- s a ed by he esiduals depic ed in he lowe pa o his igu e, whe eas a mono-exponen ial decay cu e does no desc ibe he esul s p ope ly. The wo li e imes ob ained (see Table 1) can be assigned o wo complex con igu a ions. We assume ha he bina y complex 1B N/␤-CD (deno ed as (1)) shows he sho e li e ime τ1o 0.13±0.01ms. P esumably, in addi ion 1B N complexes wi h wo cyclodex ins, such as ␤-CD/1B N/␤-CD (2) a e o med, in which he iple s a e o 1B N is somewha be e p o ec ed so ha i s li e ime τ2is subs an iallyhighe ,i.e.0.83±0.06ms. The p e-exponen ial weigh ac o s A1and A2in Table 1 poin o a a io o 2:1 o complexes (1) and (2). Also o he longe -li ed phospho escence decay cu es ob ained in he p esence o MT a bi-exponen ial i was needed; a mono-exponen ial i clea ly did no esul in an- domly dis ibu ed esiduals (see Fig. 4). In e es ingly, he longe li e imes we e signi ican ly di e en o he wo enan- iome s, i.e. 3.71±0.06 and 4.28±0.06ms o (+)-MT and (−)-MT, espec i ely. The sho e li e imes we e iden ical in bo h suspensions, i.e. 0.94±0.06 and 0.95±0.06ms, e y close o ha o s uc u e (2), measu ed in absence o MT (see Table 1). Al hough wo exponen ials a e su icien o explain he expe imen al da a o he MT complexes, i should be no ed ha a mino con ibu ion om a sho -li ing compo- nen – expec ed i complex (1) is s ill p esen – canno be ully excluded. O he species con ibu ing o he phospho escence signal (Fig.2,solidlines), hedomina ingoneis hemos in e es ing since i shows 15% RTP li e ime di e ences o he (+)-MT and he (−)-MT complexes. We a ibu e hese long-li ing signals o e na y complexes MT/1B N/␤-CD (3), whe e MT Fig. 2. Phospho escence emission spec a o 1B N/␤-CD suspension in p esence o MT enan iome s, in ai -equilib a ed suspensions ( ull lines) and a e deoxygena ion (dashed lines). Concen a ions 5×10−3M␤-CD, 5×10−5M 1B N and 5×10−4M o (+)- o (−)-MT. The lowe spec um was eco ded in absence o MT unde ae a ed condi ions. Spec ome e se ings – delay ime: 0.10ms, ga ing ime: 5.00ms; λexc =285nm, exci a ion and emission band wid hs: 10nm. C. Ga c´ıa-Ruiz e al. / Talan a 66 (2005) 634–640 637 Fig. 3. Exponen ial decay cu es and esidual analysis o phospho escence li e imes o he 1B N/␤-CD suspension unde non-deoxygena ed condi ions. p o ides p o ec ion o he 1B N exci ed iple s a e agains oxygen quenching. Mos p obably, he s e eochemical s uc- u e o (−)-MT p o ides a be e p o ec ion a e complexa- ion wi h he chi al ␤-CD han in he case o (+)-MT. I his in e p e a ion is co ec , i would be expec ed ha unde deoxygena ed condi ions, whe e he bimolecula quenching ac ion o oxygen is insigni ican , he RTP in ensi- ieso bo henan iome iccomplexeswouldbecomeiden ical. As shown in Fig. 2 (uppe cu es) his is almos ue: he e- maining di e ence in in ensi ies is only abou 4%. This may poin o mino di e ences in complex o ma ion cons an s (1)(3) o (+)-MT and (−)-MT, in ag eemen wi h he di e en p e-exponen ial ac o s A3in Table 1 and wi h he di e en complex o ma ion cu es as will be shown below. F om he ela i e inc eases in phospho escence in Fig. 2 upon deoxygena ion we can es ima e he li e ime τ0in ab- sence o quenche and he oxygen quenching a e cons an s. I shouldbe ealized ha he ex aphospho escenceis mainly due o an inc eased con ibu ion om complex (2), since i s phospho escence li e ime will be mo e s ongly enhanced Table 1 Phospho escence li e imes (τi) and weigh ac o s (Ai) o he 1B N/␤-CD suspension in absence and p esence o MT enan iome s, ob ained by i ing he expe imen al da a o a bi-exponen ial decay cu e: y=A1exp(−x/τ1)+A2exp(−x/τ2) Sys em (1) 1B N/␤-CD (2)␤-CD/1B N/␤-CD (3) MT/1B N/␤-CD A1τ1(ms) A2τ2(ms) A3τ3(ms) 1B N/␤-CD (R2=0.9998) 3.15±0.06 0.13±0.01 1.54±0.05 0.83±0.06 (+)-MT/1B N/␤-CD (R2=0.9998) 5.87±0.31 0.94±0.06 11.92±0.33 3.71±0.06 (−)-MT/1B N/␤-CD (R2=0.9998) 5.43±0.27 0.95±0.06 13.84±0.28 4.28±0.06 638 C. Ga c´ıa-Ruiz e al. / Talan a 66 (2005) 634–640 Fig. 4. Exponen ial decay cu es and esidual analysis o phospho escence li e imes o he 1B N/␤-CD suspension in he p esence o MT enan iome s unde non-deoxygena ed condi ions. C. Ga c´ıa-Ruiz e al. / Talan a 66 (2005) 634–640 639 han ha o he be e p o ec ed complex wi h MT. Using he e minology o Table 1 and ealizing ha he o al in ensi y is p opo ional o Aiτi, we can w i e: (+)-MT (53% inc ease) : 1.53(A2τ2+A3τ3)=A2τ0+A3τ0(1) (−)-MT (30% inc ease) : 1.30(A2τ2+A3τ3)=A2τ0+A3τ0(2) τ=1 τ−1 0+kO2[O2](3) in which we assume ha he phospho escence li e ime τ0(in absence o quenche ) will be he same o he complexes (2) and (3). Inse ing he p e-exponen ial ac o s and li e imes om Table 1 we ob ain o bo h (+)-MT and (−)-MT a alue o 4.3ms o τ0, compa ed o 3.71 and 4.28ms obse ed o hese complexes unde ae a ed condi ions. I ollows ha in he case o he (+)-MT complex oxygen quenching plays a ole in ae a ed suspensions: he co esponding quenching cons an can be eadily calcula ed by assuming ha he oxy- gen concen a ion is abou 10−4M: om Eq. (3) and a li e- ime o 3.7ms we ob ain a quenching a e cons an kO2o 4×105M−1s−1.This alueis3–4o de so magni udelowe han in no mal liquid solu ions, bu none heless no negligi- ble. In he case o he (−)-MT complex, howe e , oxygen quenching is no signi ican and he obse ed longes li e- ime is p ac ically equal o τ0, appa en ly in his complex he phospho opho e is much be e p o ec ed and no signi ican dec ease in li e ime is obse ed upon ae a ion. As ega ds possible analy ical applica ions o he abo e esul s, i should be no ed ha he RTP in ensi y does no a y linea ly wi h he men hol concen a ion. This should be expec ed since in he p esence o oxygen mos ly he ac ion o MT p esen in he e na y complex (3) will con ibu e o he signal. The ela i e le els o he complexes (1), (2), and (3) will change wi h he MT concen a ion, esul ing in an S- ype cu e, as depic ed in Fig. 5. Appa en ly, while a an MT concen a ion o 5×10−4M he concen a ion o he e na y complex (3) is almos a i s maximum, a an MT le el o 1×10−4M i s ela i e con ibu ion is only mino because o dissocia ion.The wocu esshowa sligh shi ,inag eemen wi h he (−) enan iome o ming a sligh ly s onge e na y complex wi h 1B N and ␤-CD han he (+) enan iome . The RTP li e ime di e ence unde ae a ed condi ions en- ables he di ec de e mina ion o he MT s e eochemis y. Whendealingwi hmix u es,weha e oconclude omFig.5 ha enan iome ic a ioso (+)-and(−)-MTmix u escanonly de e mined by RTP i he concen a ion o he leas abundan enan iome is a leas 3×10−4M. A lowe concen a ions dissocia iono he e na ycomplexeswillhampe hedis inc- iono heli e imes.The angeo enan iome ic a ios ha will be measu able depends on he p ecision o he decay cu e decon olu ion;unde hep esen condi ions heunce ain yin Fig. 5. Phospho escence in ensi y emi ed by 1B N in ␤-CD suspension as a unc ion o he men hol concen a ion. Concen a ions o ␤-CD and 1B N as usual (5×10−3and 5×10−3M, espec i ely). heli e ime(0.06ms) co esponds wi hanenan iome ic a io o 10%. Ra he unexpec ed, an in e es ing op ion o u he imp o emen migh be o ca y ou he measu emen s unde oxygen sa u a ed condi ions ins ead o no mal, ai -sa u a ed condi ions. In ha case, he oxygen concen a ion in he sus- pension will be enhanced oughly ou - old and he li e ime di e ence be ween he (+)-MT and (−)-MT complexes will become mo e p onounced. 4. Conclusions Men hol enan iome s we e ound o be e y e ec i e in- duce s o RTP o 1B N in ␤-CD aqueous suspension unde non-deoxygena ed condi ions. Using his sys em i is pos- sible o disc imina e be ween he enan iome s o his non- abso bingcompoundbydi e encesinphospho escenceli e- imes o he e na y complexes. The me hod will p obably no be sui able o ace analysis, bu ne e heless i could be ex emely use ul o samples wi h highe men hol con- cen a ions, o ins ance in he ag ance/ ood/pe sonal ca e indus y. The enan iome ic pu i y ha can be handled is s ill limi ed,andinanaly icalp ac iceachi alsepa a ions epmay s ill be necessa y. The combina ion o such a echnique wi h spec oscopicdisc imina ioniswellwo hexplo ing.Simila app oaches (as p esen ly demons a ed o men hol) can be applied o disc imina e be ween he enan iome s o o he e - penes o s uc u ally ela ed compounds wi h a cyclohexane o cyclopen ane ing s uc u e. Acknowledgmen s Ca menGa c´ ıa-Ruizg a e ully hanks heEu opeanCom- mission o a pos doc o al Ma ie Cu ie indi idual ellowship (Con ac No. HPMF-CT-2002-01826). XueShan Hu hanks 640 C. Ga c´ıa-Ruiz e al. / Talan a 66 (2005) 634–640 he China Schola ship Cen e (CSC) and NUFFIC o his g an . Re e ences [1] A. Ponce, P.A. Wong, J.J. Way, D.G. Noce a, J. Phys. Chem. 97 (1993) 11137. [2] G.M. Escanda , A.M. de la Pena, Appl. Spec osc. 55 (2001) 496. [3] X.Z. Du, Y. Zhang, Y.B. Jiang, X.Z. Huang, G.Z. Chen, Spec ochim. Ac a 53 (1997) 671. [4] H.R. Zhang, Y.S. Wei, W.L. Jin, C.S. Liu, Anal. Chim. Ac a 484 (2003) 111. [5] G.-R. Li, J.-J. Wu, W.-J. Jin, J.-W. Xie, Talan a 60 (2003) 555. [6] X. Du, Y. Zhan, Y. Jiang, X. Huang, G. Chen, Talan a 44 (1997) 511. [7] G.M. Escanda , M.A. Bold ini, Talan a 53 (2001) 851. [8] M. San os, G.M. Escanda , Appl. Spec osc. 55 (2001) 1483. [9] V.B. Naza o , V.G. A akyan, M.V. Al imo , T.G. Ve shinniko a, Russ. Chem. Bull. 52 (2002) 916. [10] Y.-Q. Zhai, S.-Z. Zhang, J.-W. Xie, C.-S. Liu, Anal. Chim. Ac a 494 (2003) 71. [11] W.J. Jin, J.M. Cos a-Fe n´ andez, A. Sanz-Medel, Anal. Chim. Ac a 431 (2001) 1. [12] B. San Vicen e de la Ri a, J.M. Cos a-Fe n´ andez, W.J. Jin, R. Pe ei o, A. Sanz-Medel, Anal. Chim. Ac a 455 (2002) 179. [13] Fe n´ andez-A g¨ uelles, J.M. Cos a-Fe n´ andez, R. Pe ei o, A. Sanz- Medel, Anal. Chim. Ac a 491 (2003) 27. [14] C. Ga c´ ıa-Ruiz, M.J. Schol es, F. A iese, C. Gooije , Talan a 66 (2005) 641. [15] N.J. Tu o, J.D. Bol , Y. Ku oda, I. Tabushi, Pho ochem. Pho obiol. 35 (1982) 69.