Fe-pillared clay mineral-based formulations of imazaquin for reduced leaching in soil

Fe-pUla ed clay mine al-based o mula ions o imazaquin o educed leaching in soil T. Undabey ia M.e. Galán-Jiménez , E. Gómez-Pan oja , J. Vázquez ' , B. Casal , F. Be gaya , E. Mo illo Slow elease o mula ions oí" he he bicide imazaquin (2-( 4-isop oPYI-4-me hyl-5-oxo-4,S-dihyd o-IH- imidazol-2-yl)quinoline-3-ca boxylic acid) we e p epa ed by i s adso p ion 00 Fe-pilla ed c1ay mine als (Fe PILCs). Fe PILes we e syn hesized by he eac ion o Na+ -mon mo illoni e (SWy-2) wi h base-hyd olyzed 50- lu ions o Fe and AL The Fej(Fe + Al) a ios used we e 0, 0.05, 0.15 and 0.50. Fe PILes showed basal spacing alues o 5.2 and 1.8 nm which we e due o ¡ on c1us e s be ween delamina ed c1ay laye s, and in e cala ed alu- minum polyoxoca ions, espec i ely.As ¡ on con en inc eased, he pilla ing p ocess yielded PILe wi h lowe mi- c opo osi y and la ge me sopo asi y. Ihe a ini y o imazaquin on Fe PIlCs was e y high ela i e o Al PIlC, as e ealed in i s highe he bidde adso p ion alues and lowe deso p ion pe cen s. Compe i i e adso p ion o an- ions such as sul a e, phospha e and chlo ide p a ided e idence o o ma ion o inne sphe e complexes o imazaquin on Fe PIlCs. Slow elease o mula ions we e p epa ed by enhanced adso p ion o he he bicide am imazaquin-cyc1odex in (CD) complexes in solu ion. COs we e able o enhance up o 8.5- old he solubili y o imazaquin, by he o ma ion o inc1usion complexes whe e he quinolinic moie y o he he bicide was pa ially apped wi hin he CO ca i y. Release o he he bidde in sandy soil was abou lj2- old lowe am Fe PIlC o mu- la ions ela i e o he comme cial o mula ion. 1. In oduc ion He bicides a e applied o impede he g ow h o weeds and im- p o e he yield o c ops. Howe e , only a pa o he applied amoun eaches i s a ge whe eas he es emains ei he a ached o he soil colloids o unde goes deg ada ion and ans e p ocesses such as mi- g a ion o he he bicide o su ace and g oundwa e , which has been epo ed o be one o he main sou ces o pollu ed wa e s (Finizio e aL, 2011; Me cal e e aL, 2011 ). Io minimize his e ec , esea ch has ocused in he las decades on he de elopmen o slow elease o - mula ions (SRFs) o he bicides ha pe mi o dec ease he losses o he used amoun s bu o achie e good bioac i i y, hus p o iding sa e en i onmen al o mula ions (Femández-U usono e aL, 2000; Sopeña e al., 2005; Undabey ia e al., 2000, 2003, 2010 ). Clay mine als ha e been s udied as good candida es o SRF because o hei good adso ben p ope ies in addi ion o hei abundance and low-cos . Mos o he s udies we e pe o med on mon mo illoni e (M ) whose su ace was modi ied p e iously om hyd ophilic o hy- d ophobic by p e-adso bing o ga üc ca ions, which we e u he used o enhanced adso p ion o he ac i e ing edien . In gene al, high adso p ion is usually co ela ed wi h slow elease p ope ies (Mishael e aL, 2003 ). Recen ly, hyd o alci es we e used based on he same p inciple, bu using o ganic anions (Zhenlan e aL , 2009 ). O he clay mine al-based app oaches we e apping o he he bicide by coagula- ion o delamina ed clay mine allaye s (Nennemann e aL , 2001a ), ad- so p ion a e p e ious he mal ea men (Bojemuelle e al , 2001 ), ben oni e polyme gels ( i e aL, 2009 ), and sepioli e gels (Maqueda e al., 2008, 2009 ). Pilla ed clay mine als (PILCs) we e p o ed e y e ec i e in he ad- so p ion o he bicides and o ganic con aminan s due o geome ical compa ibili y be ween he s uc u al po es and molecula dimensions o he chemicals, in addi ion o hei speci ic in e ac ions wi h he pil- la s and laye s o he clay mine al (Cheknane e aL, 2010 ). Adso p ion capaci y o PILC was used o he p epa a ion o SRF o he bicides (Ge s l e aL , 1998; Nennemann e aL , 2001b ). Imazaquin (IMZQ) is a selec i e imidazolinone he bicide used o b oad-spec um weed con ol in legume aops (Ab ens, 1994 ). IMZQ has an acidic ca boxyl and a basic quinoline unc ional g oup wi h pKa alues o 3.8, and 2.0, espec i ely (Regi ano e aL, 2005 ). The e o e, i ex- is s as an anion a ypical soil pH alues, inc easing i s isk o leaching down he soil p o ile. A PILC o nula ion was designed o educe i s leaching in soil (Polubeso a e al. , 2002 ). In he cu en wo k, SRFs o he he bicide imazaquin we e designed based on PILC p epa ed om hy- d olyzed i on solu ions ins ead. Imidazolinone he bicides a e known o o n complexes wi h ansi ion me als ( E e e aL, 1998 ). Imazaquin adso p ion on soils wi h di e en physicochemical p ope ies was g ea - e on he soil con aining highe amo phous i on oxide con en (Undabey ia e al., 2004). This poin ed o he o ma ion o s onge com- plexes wi h i on o e o he elemen s as aluminum, and hence, enhanced imazaquin adso p ion and s onge e en ion on Fe PILC. The alidi y o his app oach o he p epa a ion o SRF was examined in he cu en pape . In gene al, he ac i e ing edien loading on clay-based o mula ions is limi ed by he he bicide solubili y, which p e en s u he adso p ion. Cyclodex ins (CDs) a e cyclic oligosaccha ides p oduced by he enzy- ma ic b eakdown o s a ch by bac e ia. These molecules ha e a o oidal shape, wi h a hyd ophobic in e io ca i y and hyd ophilic aces (Szej li, 1998). The abili y o hese molecules o o m inclusion complexes wi h non-pola molecules has been exploi ed in he pha maceu ical indus y as a solubili y enhancemen agen (Bikia is, 2011). En i onmen al ap- plica ions o subsu ace soil emedia ion o o ganics ha e s a ed o be desc ibed based on he e y poo CD adso p ion on soil colloids (Feny esi e al., 2011; Villa e de e al., 2005). The e o e, he ac i e in- g edien con en o he PILC o mula ions can be inc eased by pe o ming adso p ion o he he bicide in solu ion om CD inclusion complexes. The he bicide will be solubilized and adso bed on he clay mine al whe eas he CD will emain mos ly in solu ion. The objec i es o his wo k we e: (i) o p epa e Fe PILCs as adso - ben s o he p epa a ion o SRF; (ii) o s udy he o ma ion o inclu- sion complexes o IMZQ wi h CD; (iii) o p epa e slow elease o mula ions om CD–IMZQ solu ions; and finally (i ) o es hese o mula ions o slow elease. 2. Ma e ials and me hods 2.1. Ma e ials Wyoming Na + –M (SWy-2) was ob ained om he Sou ce Clays Re- posi o y o The Clay Mine als Socie y (Columbia, MO) (ca ion exchange capaci y 0.8 mmol/g). Al (NO) 3 ∙9H 2 O, FeCl 3 ∙6H 2 O, NaCl, Na 2 SO 4 , Na 3 PO 4 ,H 3 PO 4 and NaOH we e pu chased om Sigma-Ald ich (Sigma Chemical Co., S . Louis, MO). HPLC g ade-ace oni ile was ob ained om Teknok omaS.A. (Ba celona, Spain). Imazaquin (IMZQ 97% pu i y) was supplied by Sigma-Ald ich and i s comme cial o mula ion (Scep e , 180 g a.i. L −1 ) by Agan Mak eshim (Lé ida, Spain). Cyclodex- ins (CDs) we e pu chased om Cyclolab (Budapes , Hunga y). CDs employed we e: α-CD, β-CD, hyd oxyp opyl-β-CD (HPβ), and andomly me hyla ed-β-CD (RAMEβ). Fig. 1 shows he s uc u al o mulas o IMZQ and CD. The uppe pa (0–20 cm) o sandy soil classified as Typic Xe opsammen was collec ed, and passed h ough a 2 mm sie e be o e use. This soil has a pH o 8.9 and 0.79% o o ganic ma e (mo e de ailed physicochemical p ope ies a e gi en in Undabey ia e al., 2012). 2.2. P epa a ion o pilla ed clay mine als Aluminum ni a e and i on chlo ide solu ions in se e al Fe/(Al + Fe) mola a ios (0, 0.05, 0.15, 0.5) we e hyd olyzed by i a ion wi h 0.4 M NaOH unde ni ogen a mosphe e. The Al/OH mola a io was 2.0. These pilla ing solu ions we e aged o 7 days. Then, 10 g o clay mine al was dispe sed in 250 mL wa e , and he pilla ing solu ions in a a io o 10 mmol (Fe + Al)/g clay mine al we e added slowly wi h a pe is al ic pump unde ni ogen a mosphe e. The final clay mine al con en was 1% (w:w). The dispe sions we e unde shaking o 24 h, ollowed by cen i uging, emo al o chlo ide by dialysis, and d y- eezing. Finally, he powde clay mine als we e hea ed o 3 h a 300 °C unde ni ogen a mosphe e. A nomencla u e was used o he PILCs whe e he fi s le e s, Al o Fe, espec i ely indica ed he absence o p esence o i on in he pilla ing so- lu ions, and he ollowing numbe s indica ed he used Fe/(Al + Fe) a io. 2.3. Cha ac e iza ion o PILC The chemical composi ion o he PILC was de e mined by X- ay fluo escence (PANaly ical Model Axios). O OR RO OR O O RO OR OR O O RO OR RO O O OOR OR RO O OR RO RO O O OR RO OR O n α-CD, n = 1, R = H β-CD, n = 2, R = H HP-β, n = 2, R = H o CH2-CHOH-CH3 RAMEβ, n = 2, R = H o CH3 Fig. 1. S uc u al o mulas o he he bicide imazaquin and cyclodex ins. 383T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389 X- ay di ac ion o he clay mine al samples was pe o med on a Philips X'Pe di ac ome e (model An on Paa HTK) wi h Cu Kα, 26 mA and 36 kV, coun ing ime o 10 s/s ep collec ed om 1 o 25° (2θ). Su ace a ea alues we e de e mined by using he BET me hod. The adso p ion o N 2 was measu ed wi h a mul i-s a ion high speed gas so p ion analyze (Quan ach ome Ins umen s NOVA 4200). The samples we e ou gassed by hea ing a 200 °C o 2 h. Fou ie ans o m in a ed (FTIR) spec a o he Fe-0.5 PILC we e eco ded in KB pelle s (2 w .% sample) using a JASCO spec ome e (model FT/IR 6300) wi h a DTGS de ec o , in he ange o 4000– 700 cm −1 . Resolu ion was o 2 cm −1 . 300 scans we e accumula ed o imp o ing he signal o noise a io in he spec a. 2.4. Adso p ion–deso p ion s udies IMZQ adso p ion expe imen s we e pe o med in duplica e by mixing 24 mg o he PILCs wi h 15 mL o he bicide solu ions anging up o 160 μM. P elimina y kine ic s udies indica ed ha adso p ion eached equilib ium a e 24 h; hus, he dispe sions we e shaken o 24 h a 20 °C. Subsequen ly, he dispe sions we e cen i uged and he concen a ion o he bicide in he supe na an s was analyzed. Deso p- ion expe imen s we e also pe o med o ce ain concen a ions (16, 64 and 160 μM) by eplacing he whole supe na an o he adso p ion expe imen s wi h wa e , e-equilib a ing o 24 h, cen i uging and u - he analyzing o he he bicide in he supe na an . Adso p ion o IMZQ on PILCs was also s udied in he p esence o NaCl, Na 2 SO 4 and Na 3 PO 4 elec oly es. The added amoun o bo h he - bicide and elec oly e anion was 160 μeq/L. Clay mine al concen a- ion was also 1.6 g/L. IMZQ–CD complexes we e p epa ed om solu ions o 100 mM CD con aining dissol ed amoun s o IMZQ as de e mined p e iously om he solubili y diag ams (1.61 and 1.29 mM o HPβand RAMEβ, e- spec i ely), by successi e dilu ions. Fe-0.5 PILC was ea ed wi h he IMZQ–CD solu ions a 1.6 g/L, and he adso bed he bicide was de e - mined as p e iously indica ed. The amoun o CD adso bed was de e - mined by analysis o he supe na an as desc ibed in Takeuchi and Miwa (1999). FTIR spec a o he he bicide be o e and a e adso p ion on Fe-0.5 PILC we e also eco ded. 2.5. Imazaquin solubili y s udies in he aqueous phase in he p esence o di e en CDs Solubili y s udies we e ca ied ou acco ding o he me hod e- po ed by Higuchi and Conno s (1965). IMZQ (10 mg) was added in excess o i s solubili y a 25 °C (60 mg/L) o aqueous solu ions (10 mL) con aining a ious concen a ions o one o he CDs. The ange was om 0 o 16 mM o β-CD and om 0 o 100 mM o α-CD, HPβand RAMEβ. The expe imen s we e ca ied ou in ipli- ca e. Flasks con aining he dispe sions we e sealed and shaken o 1 week. This eac ion ime was chosen a e p elimina y kine ic s ud- ies (da a no shown). Dispe sions we e hen fil e ed using a sy inge h ough a 0.22-μm Millipo e cellulose glass fibe memb ane fil e , and he concen a ion o IMZQ in he fil a e was de e mined by HPLC (Shimadzu Model 10A) equipped wi h a PDA de ec o as de- sc ibed in Undabey ia e al. (2008). The appa en s abili y cons an Kc was calcula ed om he s aigh line ob ained in he phase solubili y diag am, ollowing he equa ion p oposed by Higuchi and Conno s (1965): Kc ¼Slope=S01−SlopeðÞ whe e S 0 is he IMZQ equilib ium concen a ion in aqueous solu ion in he absence o CD, and Slope is he slope o he phase solubili y di- ag am. Solubili y e ficiency (Se) was also ob ained om he solubili y diag ams, which is defined as he inc emen in he appa en solubili y o IMZQ a a fixed CD concen a ion wi h espec o i s solubili y in he absence o CD. The inclusion complex o he he bicide wi h RAMEβwas cha ac- e ized by 1 H-NMR. The expe imen s we e un a 25 °C using a B uke AMX 500 spec ome e . The concen a ions employed we e 100 mM o CD and 1.28 mM o IMZQ. 2.6. P epa a ion o imazaquin o mula ions The o mula ions we e p epa ed by adso p ion o IMZQ–CD com- plexes on Fe-0.5 PILC a 1.6 g/L. The used CDs we e HP-βand RAMEβ. The used concen a ions we e 100 mM o bo h CDs, and he amoun o IMZQ dissol ed in he CD solu ions was 1.61 mM o HPβand 1.29 mM o RAMEβ. The dispe sions we e shaken o 24 h a 20 °C. Then, he he bicide was analyzed in he emo ed supe na an , whe eas he pelle s we e d y- ozen yielding he PILC o mula ions. They we e deno ed as Fe-0.5 PILC-HPβand Fe-0.5 PILC-RAMEβ. 2.7. Release o he bicide The elease o he he bicide om Fe-0.5PILC and comme cial o - mula ion was conduc ed by using Büchne unnels. In his p ocedu e 98.9 g o he sandy soil was added o a Büchne unnel (9.5 cm in e - nal diame e ) equipped wi h a pape fil e on he bo om. The soil laye was homogenized o a 0.5 cm heigh . The soil su ace was uni- o mly sp ayed wi h he di e en he bicide o mula ions a a a e o 140 kg o ac i e ing edien s pe ha. The soil laye in each unnel was i iga ed 14 imes wi h 15 mL, each washing co esponding o 2.12 mm ain a 20 min in e als. The olume elu ed a e each i iga- ion was collec ed, and he concen a ion o he bicide in he eluen was de e mined. 3. Resul s and discussion 3.1. PILC cha ac e iza ion Table 1 shows he clay mine al elemen al composi ion ob ained om he aw clay and a e pilla ing wi h Al and Al–Fe oligome s. In- e cala ion o hese oligome s was success ul as no ed in he inc ease in Al con en ela i e o he ini ial clay mine al and also in he in- c ease in Fe which pa alleled he Fe/(Al + Fe) a io used in he pilla ing p ocess. The e is also a dec ease in he oc ahed al Mg con en in he pilla ed clay mine als due o he a ack o p o ons om he oligome ic ca ions, which a e decomposed a e hea ing in oxide pilla s wi h he subsequen p o on elease (B adley and Kydd, 1993). This Mg elease yielded hyd oxyl acid si es. Howe e , hese acid si es a e no espon- sible o he s ong B öns ed acidi y epo ed in hese ypes o ma e- ials, which is a he a ibu ed o Al (o Al–Fe) pilla s (Zubko e al., 1994) as a consequence o he g ea e dissocia ion o pilla s by hy- d a ing wa e molecules (Schu z e al., 1987). Table 1 Chemical analysis o he aw clay mine al and PILCs exp essed as oxide pe cen age. SWy-2 Al-PILC Fe-0.05-PILC Fe-0.15-PILC Fe-0.5-PILC SiO 2 60.08 56.94 47.95 45.49 34.36 Al 2 O 3 17.89 21.28 24.94 23.00 16.14 Fe 2 O 3 3.84 1.00 7.39 15.00 31.18 K 2 O 0.46 0.21 0.34 0.35 0.27 MgO 2.48 2.56 1.79 1.61 1.10 CaO 1.22 0.06 0.06 0.07 0.07 Na 2 O 1.39 0.14 0.19 0.18 0.12 384 T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389 3.2. X- ay di ac ion The X- ay di ac og ams co esponding o he PILC by in e cala- ion o Al and Fe oligome s a e shown in Fig. 2. The Al PILC showed basal spacing a 1.8 nm, indica ing a 0.9 nm in e laye space in good ag eemen wi h he size o he Keggin ca ion. In con as , Al–Fe PILC showed an addi ional peak a 5.2 nm, which emained cons an when inc easing Fe/(Fe + Al) a io. This peak a low angle was o lowe in ensi y compa ed o ha a 1.8 nm o an Fe/(Al + Fe) a io o 0.05. Howe e , he con a y endency was obse ed a la ge a ios, a no iceable dec ease in he in ensi y o he peak a 1.8 nm which was shi ed o highe angles o an Fe/(Al + Fe) a io o 0.5. This endency was al eady obse ed by Mandalia e al. (1998). These au ho s ob ained basal spacing alues as high as 7.2 nm o an Fe/(Al + Fe) a io o 0.5. The di e en obse ed basal spacing alues a low angles can be due o he di e en (Fe + Al)/OH a ios used in he o ma ion o he Fe and Al oligome s, since he pilla ing p ocess is s ongly de- penden on expe imen al condi ions such as empe a u e and pH, and also dependen on he na u e o he exchangeable ca ion and es- pecially on he me al/base a io used. The high basal spacing ob ained in he cu en s udy di e ed wi h he as majo i y o p e ious s udies pe o med wi h Fe-based PILCs (Bandosz and Cheng, 1997; Cañiza es e al., 1999; Heylen and Vansan , 1997) which showed he la ges basal spacing o abou 2.5 nm. This can be due o he ac ha he s a ing ma e ials we e clay mine als wi h highe CEC han SWy-2. Al polyoxoca ions a e mo e e ec i e in neu alizing nega i e cha ge because o hei mo e plana shapes han hose o Fe ha ha e a mo e sphe ical shape (Oades, 1984). Consequen ly, he lowe he clay mine al laye cha ge, he lowe he Al-oligome compe i ion and he la ge he Fe polyoxoca ion in e cala ion. Clina d e al. (2003) obse ed by ansmis- sion elec on mic oscopy ha Al–Fe PILCs a e composed by pa icles sepa a ed om each o he by egula mesopo es, which con ained com- pac agg ega es o Fe oxide nodules. The sphe oid clus e s o med by condensa ion o hese Fe nodules had an app oxima e diame e o 3.8 nm. This alue co ela ed qui e well wi h an in e laye space o ap- p oxima ely 4.2 nm co esponding o he basal spacing o 5.2 nm, in he cu en s udy. The di ac ion obse ed a 1.8 nm was he only peak o samples p epa ed in he absence o i on; he e o e, his peak was due o he o - ma ion o Al pilla s as usually desc ibed in Al-PILC (Be gaya e al., 2006). The in ensi y dec ease obse ed o his di ac ion peak o 1.4 nm in Fe-0.5 PILC indica ed he p esence o smalle Al oligome s. The XRD da a ag eed wi h p e ious Mössbaue esul s ha indica ed no Al sub- s i u ion in he Fe pilla s and ha Al and Fe o med sepa a e pilla s o phases in he in e laye space (Aouad e al., 2010). 3.3. Po osi y The pilla ing p ocess inc eased conside ably he o al su ace a ea (Table 2) compa ed o he S BET o he aw clay mine al (32 m 2 g −1 ). The S BET dec eased sligh ly (abou 20% educ ion) wi h inc eased Fe con en . Howe e , he mic opo e a ea dec eased by abou 80% om Al PILC o Fe-0.5-PILC. The la ge dec ease o he mic opo e a ea s. he o al su ace a ea indica ed ha mos o he new su ace a ea c e- a ed in he pilla ing p ocess wi h Fe species co esponded o meso- and mac opo es. In gene al, he ma e ials p oduced om in e cala- ion o i on oxyhyd oxides c ea e smalle mic opo e olumes and su ace a eas compa ed o Al PILC (Bakas e al., 1994; Mish a e al., 1996; Pinna aia e al., 1983). This was usually a ibu ed o i on spe- cies adso bed on he ex e nal su ace o he smec i e leading o spa ial a angemen wi h po e o ma ion o di e en sizes in he mic o- and mesopo e egion (Bandosz and Cheng, 1997; Ma co-B own e al., 2012). The mesopo es esul ed om in e pa icle spaces gene a ed by h ee-dimensional coagg ega ion o i on polyoxoca ions on he clay mine al su ace o ming delamina ed s uc u es, whe eas mic o- po es a ose om in e cala ion o smalle hyd olyzed i on oxides and om mic opo ous in e s ices exis ing among he i on agg ega es and clay mine al laye s (Yuan e al., 2008). 3.4. Adso p ion iso he ms IMZQ did no adso b on he aw M (SWy-2) used in he pilla ing p ocesses due o elec os a ic epulsion a he equilib ium pH be ween he nega i ely cha ged molecules o he he bicide and he nega i ely cha ged clay mine al laye s. Howe e , he he bicide adso bed on PILC, wi h inc eased adso p ion as he i on con en inc eased (Fig. 3). This ag eed wi h p e ious s udies ha epo ed enhanced IMZQ adso p ion by Al and Fe oxyhyd oxides in soils wi h low o ganic ma e con en (Genna i e al., 1998; Regi ano e al., 1997). Undabey ia e al. (2004) s udied IMZQ adso p ion in soils wi h di e en physicochemical p op- e ies, and obse ed ha he highes adso p ion occu ed on loam sil soil due o i s high con en o amo phous i on oxides. This high a fini y o he IMZQ he bicide o i on oxides would explain alues o maximal adso p ion up o ou - old la ge han ha on Al PILC. Adso p ion on Fe PILC was ela ed o he i on con en as well as o he po osi y o he samples. IMZQ adso p ion inc eased as he meso- and mac opo e su ace inc eased. The e o e, IMZQ in e ac ion wi h he pilla s in he in e laye spaces is g ea ly acili a ed. Howe e , he g ea e IMZQ e en ion on Fe PILCs o e Al PILC was p obably e- la ed o a s onge adso p ion mechanism. Al PILC showed highe ex- e nal su ace a ea alue han Fe-0.05 PILC, bu i s IMZQ adso p ion was lowe . Polubeso a e al. (2002) sugges ed ha IMZQ adso p ion on Al PILC was occu ing h ough elec os a ic in e ac ions o he an- ionic he bicide molecules wi h posi i ely cha ged si es o pilla s ha a e a ibu ed o Lewis acidi y, in addi ion o hyd ogen bonding o B öns ed acid si es wi h he ca boxyla e moie y, ni ogen and oxygen a oms o IMZQ. Bo h mechanisms p o ide less igh binding han inne -sphe e complexa ion, acili a ing IMZQ deso p ion. Deso p ion o IMZQ was pe o med by eplacing he comple e supe na an by dis illed wa e (Table 3). The deso p ion pe cen s we e ex emely high wi h Al PILC and independen o he ini ial concen a ion, bu we e la gely educed when using Fe PILC, he la ges he inc ease in 0 5 10 15 20 25 0 50 100 150 200 250 300 In ensi y 2θ Al-PilC Fe-0.05-PilC Fe-0.15-PilC Fe-0.5-PILC 5.2 nm 1.8 nm 1.4 nm Fig. 2. X- ay di ac ion o PILCs. Table 2 Su ace cha ac e iza ion o PILCs. Sample S BET (m 2 g −1 ) Ex e nal su ace a ea (m 2 g −1 ) Mic opo e a ea (m 2 g −1 ) Al PILC 213 34 179 Fe-0.05 PILC 190 19 170 Fe-0.15 PILC 195 101 94 Fe-0.5 PILC 172 135 38 385T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389 Fe con en he lowes he deso p ion pe cen s. This sugges s ha o he adso p ion mechanisms may be ope a ing on Fe PILC. To find ou he mechanisms in ol ed, IMZQ adso p ion was pe o med on Al PILC and Fe-0.5PILC in he p esence o compe i i e anions, whose adso p ion was occu ing h ough di e en mechanisms. The com- pe i i e anions we e chlo ide, which was adso bed as ou e -sphe e complex; phospha e, as inne -sphe e complex; and sul a e, which was adso bed by bo h mechanisms (Polubeso a e al., 2000). The amoun o adso bed IMZQ dec eased in he p esence o all he elec- oly es (Table 4). Howe e , he e we e ema kable di e ences depending on he PILC. The educ ion in he amoun adso bed in Al PILC was lowe in he p esence o sul a e ela i e o chlo ide. This im- plied a la ge so p ion o he he bicide on Al PILC by an elec os a ic mechanism. On he con a y, he opposi e end was obse ed o Fe PILC, whe e he educ ion in he adso bed amoun s was mo e p om- inen in he p esence o sul a e s. chlo ide. In bo h clay mine als, he la ge dec ease in he adso bed amoun s in he p esence o phospha e was due o he inc ease in he nega i e su ace cha ge when phos- pha e was adso bed; he e o e, he bicide adso p ion would dec ease (Polubeso a e al., 2002). Ca boxylic he bicide adso p ion on i on and aluminum oxides is s ongly dependen on he he bicide s uc u e. Adso p ion o 2,4-D occu ed h ough coulombic in e ac ions be ween he o ganic anion and he posi i ely cha ged oxide su aces (Clausen and Fab icius, 2001; Wa son e al., 1973). Vasude an e al. (2002) obse ed a p e - e en ial adso p ion o he he bicide quinme ac on i on oxides such as hema i e and goe hi e o e Al oxides; howe e , he he bicide 2,4-D did no show any selec i i y. These au ho s sugges ed ha o he addi ional mechanisms on quinme ac adso p ion we e ope a - ing, and a ionalized he highe p e e ence o e Fe oxides on he con- ibu ion be ween he he e ocyclic N and Fe (III) in he o ma ion o co alen bond. The he bicides IMZQ and quinme ac ha e analogous chemical s uc u es; bo h o hem a e quinoline-ca boxylic acids, sugges ing simila adso p ion mechanisms. Ma co-B own e al. (2012) obse ed simila coo dina ion o he py idinic ni ogen o he he bicide piclo am wi h he su ace i on cen e s on Fe PILC. The esul s o his s udy also poin ed ou o a p e e en ial adso p- ion mechanism o IMZQ on Fe PILCs h ough o ma ion o inne - sphe e complexes. Leone e al. (2001) a ibu ed o a ligand exchange mechanism he adso p ion o imidazolinone he bicides on he i on oxide e ihyd i e. Bo h mechanisms can be ope a ing on Fe PILC which pe mi s o explain he high he bicide a fini y compa ed o Al PILC as no iced in he high adso p ion alues and low deso p ion pe cen s. E idence o he mechanisms in ol ed in he adso p ion o he he - bicide on he Fe PILC was pa ly suppo ed by IR spec oscopy (Fig. 4). The spec a o he he bicide showed abso p ion bands a 1732 and 1694 cm −1 due o ca bonyl s e ching ib a ions o he ca boxylic and ke one moie ies, espec i ely; mul iple ing ib a ions a ising mainly om he quinoline moie y be ween 1650 and 1300 cm −1 (1643, 1563, 1487, 1450, 1402, 1365, 1315, 1290 cm −1 ); C H in-plane bending ib a ions in he 1300–1050 cm −1 ange (specifi- cally, 1268, 1209 and 1123 cm −1 ), C H ou -o -plane bending ib a- ions below 1000 cm −1 (a 982, 858, 799 cm −1 ) and C H wagging coupled wi h a ing ib a ion a 758 cm −1 (Bajpai e al., 2000; Kum u e al., 2012; Ozël e al., 2001). In he Fe PILC (Fig. 4b), he po- si ion and assignmen o he main abso p ion bands we e (Qin e al., 2010; Yuan e al., 2008): 1636 cm −1 o OH bending o wa e ; 1013 cm −1 o Si O s e ching ib a ion; 916 cm −1 o Al Al OH de o ma ion; and 882 and 796 cm −1 o Si O ib a ions o some qua z impu i ies o he aw mine al. The IR spec um a e adso p- ion o he he bicide (Fig. 4c) showed some ea u es sugges ing he in ol emen o N-dono ligand and o ma ion o inne -sphe e com- plexes: i) Fe–N coo dina ion h ough he quinoline moie y we e e- po ed by Özel e al. (2001) o yield upwa d shi s o mos o he ing ib a ions be ween 4 and 22 cm −1 as in Fig. 4c (1457, 1408, 1369, 1329, 1297 cm −1 ); and ii) he absence o he ing ib a ion bands a 1643 and 1563 cm −1 o he p is ine he bicide (Fig. 4a) which a e associa ed o p o ona ed quinoline ni ogen (quinolinium moie ies) (Pusino e al., 2003) and used as a finge p in o de e mi- na ion o B öns ed acid si es in clay mine als (Co ma e al., 1993), also sugges ed coo dina ion o he N a om o he quinoline ing o Fe cen- e s. In addi ion, he new abso p ion bands a 1250 and 1188 cm −1 we e due o C O s e ching ib a ions esul ing om he eac ion o he ca boxylic moie y wi h Fe oxyhyd oxides, simila ly o es e s o ino ganic alcohols. The lack o de ec ion o he C_O s e ching e- quency in hese ypes o in e ac ions is due o he ac ha i can a y wi hin e y wide limi s and he e o e, no ound a he ypical ca - bonyl equencies o no mal sa u a ed es e s wi h o ganic alcohols (1750–1730 cm −1 )(Bellamy, 1975). These esul s also poin ou o - ma ion o inne -sphe e complexes o he he bicide h ough i s ca - boxylic moie y o Fe pilla s. 3.5. Solubili y diag ams o imazaquin in he p esence o cyclodex ins Solubili y diag ams o IMZQ wi h ou CDs showed a linea in- c ease in he he bicide solubili y when inc easing CD concen a ion (Fig. 5). These cu es we e indica ing o ma ion in solu ion o com- plexes wi hou a defined solubili y limi , which co esponded o A L ype acco ding o Higuchi and Conno s classifica ion (Higuchi and Conno s, 1965). A L ype indica es o ma ion o 1:1 inclusion com- plexes be ween CD and he gues molecule. The condi ional o ma ion 0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 Imazaquín adso bed (µmol/g) Equilib ium concen a ion (µM) Al-PilC Fe-0.05-PilC Fe-0.15-PilC Fe-0.5-PilC Fig. 3. Adso p ion iso he ms o imazaquin on PILCs. Table 3 Pe cen ages o imazaquin deso bed as a unc ion o he concen a ion used. a IMZQ added, μM Al PILC Fe-0.05 PILC Fe-0.15 PILC Fe-0.5 PILC 16 75.8 ± 6.1% 80.7 ± 5.4% 83.3 ± 9.0% 52.3 ± 1.0% 64 73.0 ± 5.7% 80.5 ± 3.2% 66.4 ± 2.1% 48.6 ± 0.1% 160 73.4 ± 3.6% 64.5 ± 2.3% 56.6 ± 1.8% 38.5 ± 0.2% a Clay mine al concen a ion was 1.6 g/L. Table 4 Reduc ion pe cen s in imazaquin adso p ion on PILCs in he p esence o se e al elec- oly es. Imazaquin o anion a io was 1:1. a NaCl Na 2 SO 4 Na 3 PO 4 Al PILC 11.01 ± 1.63 6.18 ± 0.57 86.72 ± 1.94 Fe-0.5 PILC 30.42 ± 1.84 74.51 ± 1.37 92.58 ± 3.5 a The added amoun o he bicide was 160 μM. 386 T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389 cons an s o he complexes we e calcula ed as well as he maximal in- c emen in he he bicide solubili y (Table 5). The alues o he appa en s abili y cons an s (Kc) o β-CD and i s de i a i es we e much la ge han ha o α-CD. The low alue o his CD can be due o he mo e educed dimensions o i s ca i y, which would hinde he en ance o he he bicide; he e o e he in e ac ion mode will be di e en when using β-CD and de i a i es which p esen a mo e op imal accommoda ion be ween he hyd ophobic ca i y and he non-pola moie ies o he he bicide. The highes complexa ion cons an as well as he maximal inc e- men in IMZQ solubili y occu ed wi h HPβ(up o 8- old o 0.1 M concen a ion). A simila alue o cons an was obse ed o β-CD bu he enhanced solubili y o he he bicide was lowe due o i s lowe solubili y. The enhanced solubili y o he he bicide was also e- ma kable wi h RAMEβ, wi h solubili y inc emen s up o 6- old o a 0.1 M concen a ion and high complexa ion cons an . 3.6. S udy o imazaquin–cyclodex in inclusion complexes 1 H-NMR spec oscopy is a use ul and powe ul ool which pe mi s he cha ac e iza ion o complexes in solu ion by moni o ing he chemical shi s o CD-p o ons and hose o he gues molecule. Table 6 shows he chemical shi s o he he bicide and o i s inclu- sion complex wi h RAMEβ. The shi s o CD p o ons we e no possi- ble o moni o due o he high ee CD concen a ion ela i e o he ac ion o ming inclusion complexes. Consequen ly, he changes we e only ollowed in he he bicide spec a whose peaks did no o e lap wi h hose o he CD. S udies on inclusion complexes o β-CD and de i a i es showed shi s o low fields o he gues molecule p o ons which a e in e ac ing wi hin he CD ca i y (Goyenechea e al., 2001; Lezcano e al., 2003; Pé ez-Ma ínez e al., 2000). In Table 6, he highes shi was ob ained o H-4. High shi s we e also obse ed o H-6 and H-7. These ea u es indica ed in e ac ion o he he bicide ia he quinolinic ing wi h he CD. Howe e , he shi s o hese a oma ic p o ons we e occu ing a high field ins ead. Masson e al. (1998) obse ed a simila e ec on chlo ambucil inclusion wi h β-CD, which was explained on he basis o he opposi e e ec p oduced by s ong hyd ogen bonding in ol ing N elec o pai wi h he CD. These da a poin ed ou ha he quinolinic moie y o he he bicide is pa ially apped wi hin he CD ca i y, wi h he quinolinic N a om ou side o he ca i y and o ming s ong H-bonding wi h pola ized wa e molecules o hyd oxyl g oups o he CD. The small chemical shi s o p o ons loca ed in he imidazolinone ing and i s subs i uen g oups indica e ha his ing was s aying ou - side he ca i y. This in e ac ion was ied o find ou which CD p o ons we e in e ac ing wi h H-4 by pe o ming NOE expe imen s. Howe e , he esul s we e unsuccess ul due o he high amoun o non- complexed CDs. Simila obse a ions we e ob ained o he o he CDs. 3.7. Adso p ion o imazaquin in he p esence o cyclodex ins In Fig. 3, adso p ion o IMZQ was highes on Fe-0.5 PILC. I s ad- so p ion was linea and limi ed by he solubili y o he he bicide; he e o e, his PILC was selec ed o inc ease IMZQ adso p ion in he 1800 1600 1400 1200 1000 800 1297 753 858 916 1006 1081 1150 965 1188 1329 1250 1369 1408 1457 c) Wa enumbe (cm-1) 1636 882 796 779 916 1013 1116 b) 1636 1290 1315 734 799758 982 1209 858 1268 1123 1365 1450 1487 1402 1044 1563 1643 1694 1732 a) Fig. 4. FTIR spec a o (a) IMZQ, (b) Fe-0.5-PILC and (c) a e IMZQ so p ion on Fe-0.5-PILC. 0 20 40 60 80 100 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 Imazaquin concen a ion (mM) CD concen a ion (mM) Fig. 5. Phase solubili y diag ams o imazaquin in aqueous solu ions in he p esence o CD: ( )α-CD, (■)β-CD, ( ) RAMEβand ( )HPβ. Table 5 Appa en s abili y cons an s (Kc), co ela ion ac o s (R 2 ), and solubili y inc ease (Se, using 0–0.016 M o β-CD and 0–0.1 M o he o he CDs) o he inclusion compounds ob ained be ween imazaquin and he di e en CDs. β-CD HPβRAMEβα-CD K(M −1 ) 71.0 ± 4.7 85.4 ± 2.6 59.2 ± 2.6 11.5 ± 3.4 Se 2.0 8.5 6.3 2.0 R 2 0.960 0.990 0.981 0.966 Table 6 Chemical shi s o ee imazaquin and RAMEβ: IMZQ complexes om 1 H-NMR measu emen s (500 MHz, DMSO). P o on Signal δ(IMZQ) δ(RAMEβ–IMZQ) Δδ H-4 Single 8.927 8.845 −0.082 H-6 Double double 8.227 8.187 −0.040 H-7 Mul iple 7.972 7.927 −0.045 H-8 Mul iple 7.803 7.772 −0.031 H-9 Double double 8.169 8.144 −0.025 CH 3 Single 1.275 1.254 −0.021 CH(CH 3 ) 2 Double 1.021 1.015 −0.006 CH(CH 3 ) 2 Double 0.863 0.865 +0.002 CH(CH 3 ) 2 Mul iple 1.960 1.949 −0.011 387T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389 p esence o CD. Adso p ion o he he bicide om CD–IMZQ would p omo e he amoun o adso bed he bicide by enhancemen o IMZQ solubili y. The CD would be only ac ing as a solubilizing agen o inc ease he ac i e ing edien con en in he PILC-based o mula- ions. The CDs used we e only RAMEβand HPβbecause hei he bi- cide solubili y inc emen s a e he la ges (Table 5). Adso p ion o he he bicide om CD–IMZQ complexes inc eased la gely he amoun o he bicide adso bed compa ed o i s adso p ion in he absence o CD (Fig. 6). He bicide adso p ion inc eased by 42% in he p esence o RAMEβ. Smalle inc emen s we e ob ained o HPβ, which is in good ag eemen wi h i s highe complexa ion cons an . The CD in solu ion is compe ing wi h he so p ion si es on he Fe PILC o IMZQ, impeding a la ge adso p ion. Adso p ion o bo h CD on Fe PILC was de e mined o be negligible. 3.8. Release s udies Fig. 7 shows he elu ion cu es o he comme cial o mula ion Scep e as well as hose o Fe-0.5 PILC p epa ed in he p esence o CD. A e 14 i iga ions, he amoun eleased om he o al applied was 87.8 ± 3.1% o he comme cial o mula ion. The high elu ion ob ained in his sandy soil wi h he comme cial o mula ion ag ees wi h he s udy by Undabey ia e al. (2004) who de e mined ha ad- so p ion o he he bicide on his soil mainly occu ed h ough he o - ganic ma e , om which deso bed easily. A e 14 i iga ions, he o al amoun s eleased om he o al applied we e 40.2 ± 6.7 and 47.4 ± 4.8% o hose p epa ed, espec i ely, wi h RAMEβand HPβ, amoun ing o a significan educ ion in elease o 54 and 46%, espec- i ely, compa ed o he comme cial o mula ion. 4. Conclusions Fe PILCs we e syn hesized o hedesign o slow elease o mula ions o he he bicide IMZQ. The app oach used was he high a fini y o his he bicide obse ed o i on oxide su aces; he e o e, an enhanced ad- so p ion o he he bicide on i on-based clay mine als which pa alleled a slow elease could be achie ed. Fe PILCs we e syn hesized and cha ac- e ized. XRD and po osi y measu emen s we e in ag eemen wi h p e i- ous s udies indica ing a mesopo ous s uc u e o med by coagg ega ion o i on oxide clus e s wi h clay mine al delamina ed laye s. Unless igh binding mechanisms a e ope a ing, a la ge po e will acili a e he in e - ac ion o he he bicide wi h i on nodulesand also IMZQ deso p ion om he Fe PILCs. Compe i i e adso p ion o anions showed ha IMZQ was adso bed mainly by inne -sphe e complexa ion; howe e , o he mechanisms may be ac ing. Inne -sphe e complexa ion was also sup- po ed by IR spec oscopy. The inc ease in he loading o he bicide on clay mine als o p epa- a ion o o mula ions is usually hinde ed by he he bicide solubili y. This is o e come by he use o solubilizing agen s, usually su ac an s. Howe e , he use o CDs, which a e non- oxic, biodeg adable chemicals, is a good al e na i e. IMZQ solubili y was g ea ly enhanced in he p es- ence o CD, by he o ma ion o inclusion complexes. The use o hese complexes inc eased he amoun o IMZQ adso bed on he clay mine al up o 42%. Release o he he bicide in sandy soil om o mula ions p e- pa ed om CD–IMZQ complexes was e a ded by 1/2- old compa ed o he comme cial o mula ion Scep e . Acknowledgmen s This esea ch was suppo ed by he MEC P ojec CTM2009-07425 and he Jun a de Andalucía P ojec P09-RNM4581. Bo h p ojec s e- cei ed unding by he Eu opean Social Fund. The au ho s acknowledge he CITIUS se ice om he Uni e si y o Se ille o he NMR acili ies. Re e ences Ah ens, H.W., 1994. He bicide Handbook, 7 ed. Weed Science Socie y o Ame ica, Champaign, IL. Aouad, A., Anas ácio, A.S., Be gaya, F., S ucki, J.W., 2010. A Mössbaue spec oscopy s udy o aluminum- and i on-pilla ed clay mine als. Clays and Clay Mine als 58, 164–173. Bajpai, U.C., Gup a, D.C., Singh, R.B., Saxena, M., Saxena, A.K., 2000. Fou ie ans o m in- a ed spec a and no mal mode analysis o 1-(3-me hyl phenyl pipe azin-1-yl)-2- (quinolin-2-yl) e hane (cen haquin): a po en cen ally ac ing an i-hype ensi e agen . Jou nal o Molecula S uc u e 516, 15–21. Bakas, T., Mouka ika, A., Papae hymiou, V., Lada os, A., 1994. Redox ea men o an Fe/Al pilla ed mon mo illoni e. A Mössbaue s udy. Clays and Clay Mine als 42, 634–642. Bandosz, T.J., Cheng, K., 1997. Changes in acidi y o Fe-pilla ed/delamina ed smec i es on hea ea men . Jou nal o Colloid and In e ace Science 191, 456–463. Bellamy, L.J., 1975. The In a-Red Spec a o Complex Molecules. Chapman & Hall L d., London, GB. Be gaya, F., Aouad, A., Mandalia, T., 2006. Pilla ed clays and clay mine als. In: Be gaya, F., Theng, B.K.G., Lagaly, G. (Eds.), Handbook o Clay Science, pp. 393–421 (Chap e 7.5). Bikia is, D.N., 2011. Solid dispe sions, pa I: ecen e olu ions and u u e oppo uni ies in manu ac u ing me hods o dissolu ion a e enhancemen o poo ly wa e - soluble d ugs. Expe Opinion on D ug Deli e y 8, 1501–1519. Bojemuelle , E., Nennemann, A., Lagaly, G., 2001. Enhanced pes icide adso p ion by he mally modified ben oni es. Applied Clay Science 18, 277–284. B adley, S.M., Kydd, R.A., 1993. Ga 13 ,Al 13 ,GaAl 12 , and ch omium-pilla ed mon mo illoni es- acidi y and eac i i y o cumene con e sion. Jou nal o Ca alysis 141, 239–249. Cañiza es, P., Val e de, J.L., SunKou, M.R., Molina, C.B., 1999. Syn hesis and cha ac e iza- ion o PILCs wi h single and mixed oxide pilla s p epa ed om wo di e en ben on- i es. A compa a i e s udy. Mic opo ous and Mesopo ous Ma e ials 29, 267–281. 0 200 400 600 800 1000 1200 0 10 20 30 40 50 60 70 80 90 100 110 120 130 Imazaquin adso bed (µmol/g) Equilib ium concen a ion (µM) Fig. 6. Adso p ion iso he ms o imazaquin on Fe-0.5-PILC in he absence (■) and p esence o RAMEβ() and HPβ( ). 0 2 4 6 8 10121416 0 5 10 15 20 25 30 35 40 Imazaquin elu ed (% o he applied) Numbe o i iga ions Scep e Fe-0.5-PilC-HP-ß-CD Fe-0.5-PilC-RAMEB-CD Fig. 7. He bicide elease om he comme cial and Fe PILC o mula ions. 388 T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389 Cheknane, B., Baudu, M., Basly, J.P., Bou as, O., 2010. Adso p ion o basic dyes in single and mix u e sys ems on g anula ino ganic–o ganic pilla ed clays. En i onmen al Technology 31, 815–822. Clausen, L., Fab icius, I., 2001. A azine, isop o u on, mecoc op, 2,4-D, and ben azone adso p ion on o i on oxides. Jou nal o En i onmen al Quali y 30, 858–869. Clina d, C., Mandalia, T., Tchouba , D., Be gaya, F., 2003. HRTEM image fil a ion: nanos uc u al analysis o a pilla ed clay. Clays and Clay Mine als 51, 421–429. Co ma, A., Fo nés, V., Rey, F., 1993. Quinoline as a p obe molecule o de e mina ion o ex e nal B öns ed and Lewis acidi y in zeoli es. Zeoli es 13, 56–59. E e, L.S., Ga ibba, E., Mice a, G., Sa done, N., 1998. Me al complexes o imazapy , a he bicide p o ided wi h e ficien me al-chela ing abili y: c ys al s uc u e o he cobal (II) and manganese (II) complexes. Ino ganica Chimica Ac a 272, 68–73. Feny esi, E., Balogh, K., Olah, E., Ba ai, B., Va ga, E., Molna , M., G uiz, K., 2011. Cyclo- dex ins o emedia ion o soils con amina ed wi h chlo ina ed o ganics. Jou nal o Inclusion Phenomena and Mac ocyclic Chemis y 70, 291–297. Fe nández-U usono, R., Ginés, J.M., Mo illo, E., 2000. De elopmen o con olled e- lease o mula ions o alachlo in e hylcellulose. Jou nal o Mic oencapsula ion 17, 331–342. Finizio, A., Azimon i, G., Villa, S., 2011. Occu ence o pes icides in su ace wa e bodies: a c i ical analysis o he I alian na ional pes icide su ey. Jou nal o En i onmen al Moni o ing 13, 49–57. Genna i, M., Neg e, M., Vind ola, D., 1998. Adso p ion o he he bicides imazapy , imaze hapy and imazaquin on soils and humic acids. Jou nal o En i onmen al Science and Heal h B33, 547–567. Ge s l, Z., Nasse , A., Mingelg in, U., 1998. Con olled elease o pes icides in o soils om clay–polyme o mula ions. Jou nal o Ag icul u al and Food Chemis y 46, 3797–3802. Goyenechea, N., Sánchez, M., Velaz, I., Ma ín, C., Ma ínez-Oha iz, M.C., González- Gai ano, G., 2001. Inclusion complexes o nabume one wi h be a-cyclodex ins: he modynamics and molecula modelling s udies. Influence o sodium pe chlo- a e. Luminescence 16, 117–127. Heylen, I., Vansan , E.F., 1997. The di e ence in adso p ion capaci y be ween Fe-PILCs and modified Fe–BuA- and Fe–Z -PILCs. Mic opo ous Ma e ials 10, 41–50. Higuchi, T., Conno s, K.A., 1965. Phase-solubili y echniques. Ad ances in Analy ical Chemis y and Ins umen a ion 4, 117–212. Kum u, M., Küçük, V., Ba dakçi, T., 2012. Theo e ical and expe imen al s udies on he i- b a ional spec a o 3-quinolinca boxaldehyde. Spec ochimica Ac a Pa A 90, 28–34. Leone, P., Genna i, M., Neg e, M., Boe o, V., 2001. Role o e ihyd i e in adso p ion o h ee imidazolinone he bicides. Jou nal o Ag icul u al and Food Chemis y 49, 1315–1320. Lezcano, M., No o, M., Al-Soufi,W., Rod íguez-Núñez, E., Ta o, J.V., 2003. Complexa ion o se e al ungicides wi h be a-cyclodex in: de e mina ion o he associa ion con- s an s and isola ion o he solid complexes. Jou nal o Ag icul u al and Food Chem- is y 51, 5036–5040. Li, J.F., Jiang, M., Wu, H., Li, Y.M., 2009. Addi ion o modified ben oni es in polyme gel o mula ion o 2,4-D o i s con olled elease in wa e and soil. Jou nal o Ag icul- u al and Food Chemis y 57, 2868–2874. Mandalia, T., C espin, D., Messad, D., Be gaya, F., 1998. La ge in e laye epea dis ances obse ed o mon mo illoni es ea ed by mixed Al–Fe and Fe pilla ing solu ions. Chemical Communica ions 19, 2111–2112. Maqueda, C., Villa e de, J., Sopeña, F., Undabey ia, T., Mo illo, E., 2008. No el sys em o educing leaching o he he bicide me ibuzin using clay–gel-based o mula ions. Jou nal o Ag icul u al and Food Chemis y 56, 11941–11946. Maqueda, C., Villa e de, J., Sopeña, F., Undabey ia, T., Mo illo, E., 2009. E ec s o soil cha ac e is ics on me ibuzin dissipa ion using clay–gel-based o mula ions. Jou - nal o Ag icul u al and Food Chemis y 57, 3273–3278. Ma co-B own, J.L., Ma ín Ba bosa-Lema, C., To es Sánchez, R.M., Me cade , R.C., dos San os A onso, M., 2012. Adso p ion o piclo am he bicide on i on oxide pilla ed mon mo illoni e. Applied Clay Science 58, 25–33. Masson, M., Lo sson, T., Jóndó i , S., F idiksdó i , H., 1998. S abilisa ion o ionic d ugs h ough complexa ion wi h non-ionic and ionic cyclodex ins. In e na ional Jou - nal o Pha maceu ics 164, 45–55. Me cal e, C.D., Beddows, P.A., Boucho , G.G., Me cal e, T.L., Li, H.X., Van La ie en, H., 2011. Con aminan s in he coas al ka s aqui e sys em along he Ca ibbean coas o he Yuca an Peninsula, Mexico. En i onmen al Pollu ion 159, 991–997. Mishael, Y.G., Undabey ia, T., Rabino i z, O., Rubin, B., Ni , S., 2003. Sul osul u on inco - po a ed in micelles adso bed on mon mo illoni e o slow elease o mula ions. Jou nal o Ag icul u al and Food Chemis y 51, 2253–2259. Mish a, T., Pa ika, K.M., Rao, S.B., 1996. T ansi ion me al oxide pilla ed clay: 1. A com- pa a i e s udy o ex u al and acidic p ope ies o Fe(III) pilla ed mon mo illoni e and pilla ed acid ac i a ed mon mo illoni e. Jou nal o Colloid and In e ace Sci- ence 183, 176–183. Nennemann, A., Kulbach, S., Lagaly, G., 2001a. En apping pes icides by coagula ion smec i es. Applied Clay Science 18, 285–298. Nennemann, A., Mishael, Y.G., Ni , S., Rubin, B., Polubeso a, T., Be gaya, F., an Damme, H., Lagaly, G., 2001b. Clay-based o mula ions o me olachlo wi h educed leaching. Applied Clay Science 18, 265–275. Oades, J.M., 1984. In e ac ions o polyca ions o aluminium and i on wi h clays. Clays and Clay Mine als 32, 49–57. Özel, A.E., Büyükmu a , Y., Akyüz, S., 2001. In a ed-spec a and no mal-coo dina e analysis o quinoline and quinoline complexes. Jou nal o Molecula S uc u e 565–566, 455–462. Pé ez-Ma ínez, J.I., Ginés, J.M., Mo illo, E., Moyano, J.R., 2000. 1 H-nuclea magne ic es- onance and phase solubili y s udies o he s oichiome ies in 2,4-D: alpha- and be a-cyclodex ins inclusion complexes. Jou nal o Inclusion Phenomena and Mac- ocyclic Chemis y 37, 171–178. Pinna aia, T.J., Ray ha ha, R.H., Tzou, M.S., Landau, S.D., 1983. Pilla ed and delamina ed clay ca alys s. Abs ac s o Pape s o he Ame ican Chemical Socie y 186 (59-INDE). Polubeso a, T., Undabey ia, T., Ni , S., Che ko a, L., Van Damme, H., Annabi-Be gaya, F., 2000. Adso p ion o sul ome u on and o he anions on pilla ed clay. Jou nal o En- i onmen al Quali y 29, 948–954. Polubeso a, T., Ni , S., Ge s l, Z., Bo iso e , M., Rubin, B., 2002. Imazaquin adso bed on pilla ed clay and c ys al iole -mon mo illoni e complexes o educed leaching in soil. Jou nal o En i onmen al Quali y 31, 1657–1664. Pusino, A., Gelsomino, A., Fio i, A., Gessa, C., 2003. Adso p ion o wo quinolineca boxylic acid he bicides on homoionic mon mo illoni es. Clays and Clay Mine als 51, 143–149. Qin, Z., Yuan, P., Zhu, J., He, H., Liu, D., Yang, S., 2010. Influences o he mal p e ea - men empe a u e and sol en on he o ganosilane modifica ion o Al13- in e cala ed/Al-pilla ed mon mo illoni e. Applied Clay Science 50, 546–553. Regi ano, J.B., Bischo , M., Lee, L.S., Reiche , J.M., Tu co, R.F., 1997. Re en ion o imazaquin in soil. En i onmen al Toxicology and Chemis y 16, 397–404. Regi ano, J.B., Da Rocha, W.S.D., Alleoni, L.R.F., 2005. Soil pH on mobili y o imazaquin in oxisols wi h posi i e balance o cha ges. Jou nal o Ag icul u al and Food Chem- is y 53, 4096–4102. Schu z, A., Plee, D., Bo g, F., Jacobs, P., Poncele , G., F ipia , J.J., Van Olphen y, H., 1987. Acidi y and ca aly ic p ope ies o pilla ed mon mo illoni e and beidelli e. In: Schul z, L.C., Mump on, F.A. (Eds.), P oceedings o he In e na ional Clay Con e - ence, Den e , 1985. The Clay Mine als Socie y, Blooming on, Indiana, pp. 305–310. Sopeña, F., Cab e a, A., Maqueda, C., Mo illo, E., 2005. Con olled elease o he he bi- cide no flu azon in o wa e om e hylcellulose o mula ions. Jou nal o Ag icul u - al and Food Chemis y 53, 3540–3547. Szej li, J., 1998. In oduc ion and gene al o e iew o cyclodex in chemis y. Chemical Re iews 98, 1743–1753. Takeuchi, T., Miwa, T., 1999. Indi ec fluo ime ic de ec ion o cyclodex ins ia pos column mixing wi h 2-p- oluidinyl-6-naph halenesul ona e in liquid ch oma- og aphy. Jou nal o High Resolu ion Ch oma og aphy 22, 609–612. Undabey ia, T., Ni , S., Rubin, B., 2000. O gano-clay o mula ions o he hyd ophobic he bicide no flu azon yield educed leaching. Jou nal o Ag icul u al and Food Chemis y 48, 4767–4773. Undabey ia, T., Mishael, Y.G., Ni , S., Papahadjopoulos-S e nbe g, B., Rubin, B., Mo illo, E., Maqueda, C., 2003. A no el sys em o educing leaching om o mula ions o anionic he bicides: clay–liposomes. En i onmen al Science and Technology 37, 4475–4480. Undabey ia, T., Sánchez-Ve dejo, T., Mo illo, E., Maqueda, C., 2004. E ec o o ganic amendmen s on he e en ion and mobili y o imazaquin in soils. Jou nal o Ag i- cul u al and Food Chemis y 52, 4493–4500. Undabey ia, T., Ni , S., Sánchez-Ve dejo, T., Villa e de, J., Maqueda, C., Mo illo, E., 2008. A clay– esicle sys em o wa e pu ifica ion om o ganic pollu an s. Wa e Re- sea ch 42, 1211–1219. Undabey ia, T., Sopeña, F., Sánchez-Ve dejo, T., Villa e de, J., Ni , S., Mo illo, E., Maqueda, C., 2010. Pe o mance o slow- elease o mula ions o alachlo . Soil Sci- ence Socie y o Ame ica Jou nal 74, 898–905. Undabey ia, T., Recio, E., Maqueda, C., Sánchez-Ve dejo, T., Balek, V., 2012. Slow elease o mula ions o diu on based on clay–phospha idylcholine complexes. Applied Clay Science 55, 53–61. Vasude an, D., Coope , D.M., Van Exem, O.L., 2002. So p ion–deso p ion o ionogenic compounds a he mine al wa e in e ace: s udy o me al oxide- ich soils and pu e phase mine als. En i onmen al Science and Technology 36, 501–511. Villa e de, J., Pé ez-Ma ínez, J.I., Maqueda, C., Ginés, J.M., Mo illo, E., 2005. Inclusion complexes o α- and γ-cyclodex ins and he he bicide noflu azon: I. P epa a ion and cha ac e iza ion. II. Enhanced solubilisa ion and emo al om soils. Chemosphe e 60, 656–664. Wa son, J.R., Posne , A.M., Qui k, J.P., 1973. Adso p ion o he he bicide 2,4-D on goe- hi e. Jou nal o Soil Science 24, 503–511. Yuan, P., Annabi-Be gaya, F., Tao, Qj, Fan, M., Liu, Z., Zhu, J., He, H., 2008. A combined s udy by XRD, FTIR, TG and HRTEM on he s uc u e o delamina ed Fe- in e cala ed/pilla ed clay. Jou nal o Colloid and In e ace Science 324, 142–149. Zhenlan, Q., Heng, Y., Bin, Z., Wanguo, H., 2009. Syn hesis and elease beha iou o bac- e icides in e cala ed Mg–Al laye ed double hyd oxides. Colloids and Su aces A 348, 164–169. Zubko , S.A., Kus o , L.M., Kazansky, V.B., Fe e , G., Tichi , D., Figue as, F., 1994. S udy o he na u e o acid si es o mon mo illoni es pilla ed wi h aluminium and oligosilsesquioxane complex ca ions. 1. B öns ed acidi y. Clays and Clay Mine als 42, 421–427. 389T. Undabey ia e al. / Applied Clay Science 80–81 (2013) 382–389