The fate of Hg0 in natural waters

Several aspects of the aquatic chemistry of elemental mercury (Hg0) are vital to the understanding of the fate of this metal in the environment, yet have scarcely been studied. Reactive dissolution of metallic mercury is shown to be dependent on the metal concentration in solution, the metal oxidation rate, particulate material concentration, nature and concentration of other ions, and pH. When using 1 g L-1 of Hg0 in distilled water open to the atmosphere, the concentration of soluble mercury increases as a function of time, reaching 5.4 mg L-1 of total Hg in the steady state. From this, 3.2 mg L-1 were due to Hg2+ formed via oxidation. In lake water, results showed an inhibition in the reactive dissolution process, and the total metal concentration in water was 3.1 mg L-1 in the steady state. This inhibitory effect was attributed to particulate material. In seawater, the total concentration of soluble Hg increases as a function of time, reaching a peak of 17.8 mg L-1 after 10 h. After this, the soluble concentration dropped, to 4.8 mg L-1. Experiments performed at different values of pH (4.0;7.0 and 9.0), showed that the dissolution of the metal occurred to a higher extent at pH 4.0. Adsorption studies of both mercuric ions and elemental mercury species onto particulate material showed a dependence on the surface area, following the sequence 400 mesh > 200 mesh > sediment in natura. The implications of such findings are discussed, taking into consideration the Amazonian scenario.


Intro duc ti on
Pre dic ting the real im pact re la ted to emis si ons of mercury into the en vi ron ment still po ses many pro blems, sin ce some of the fun da men tal as pects of the aqua tic che mistry of mer cury have not yet been fully stu di ed. Mer cury spe cies re a ching the aqua tic system are, in a bro ad sen se, clas sifi ed into inor ga nic (Hg 0 , Hg 2+ and Hg 2 2+ ) and or ga nic (CH 3Hg + , (CH 3)2Hg, etc.) forms 1,2 .
Orga nic mer cu ri al com pounds are con si de red to be the most to xic form of the me tal. Due to the ir li pid so lu bi lity, the se spe ci es can quickly en ter the blo od stre am and da ma ge the cen tral ner vous system 3,4 . Inor ga nic forms of mer cury have much lo wer to xi city than the or ga nic ones, with a decre a se in to xi city from to Hg 2+ > Hg 2 2+ > Hg 0 . Re cently, Jardim et al. 5 , sho wed that in mi cro bi al acu te to xi city tests (up to 2 h), the inhi bi ti on in the res pi ra ti on of Esche ri chia coli ob ser ved when stres sed by 5000 mg L -1 of Hg 0 was si mi lar to the one ca u sed by 10 µg L -1 of Hg 2+ , sho wing the low to xicity of the ele men tal spe ci es com pa red to mer cu ric ions.
Worl dwi de, in dus tri al and agri cul tu ral uses of mer cury have drop ped mar kedly in the last three de ca des. Pre sently, most coun tri es are fa cing old in ven tory re me di a ti on ac tions rat her than de a ling with pre sent in puts into the ecosystems. Ho we ver, this is not the pre sent sce na rio in Bra zil whe re, due to the gro wing ac ti vity of gold ex trac ti on ( garim pos ) which use mer cury to amal ga ma te gold, it has been cal cu la ted that about 1,500-3,000 to nes of Hg have been rele a sed in the Ama zon re gi on in the last 15 ye ars 6 .
Me tal lic mer cury used in the ga rim pos can re ach the aqua tic system eit her by di rect spills of li quid dro plets, or by fu gi ti ve at mosp he ric emis si on du ring the ro as ting of the Au-Hg amal gam. The re si den ce time of ele men tal mer cury in the at mosp he re de pends on the con cen tra ti on of oxi dant spe ci es as well as the de po si ti on rate of mer cury va pors. Upon re a ching the wa ter bo di es, in de pen dent of the emis sion sour ce, the fate of mer cury de pends only on the physico-chemical and bi o lo gi cal fe a tu res of the wa ter body 2 . The first and most im por tant pro cess con cer ning mer cury spe ci a ti on is the par ti ti o ning bet we en so lid and aque ous pha se. In this case, mer cu ric ions should have more af fi nity for so lids than the ele men tal form and ad sorp ti on can be con si de red as a key me cha nism to ame li o ra te the me tal eco to xi co logy. Also, de pen ding on the re dox po ten ti al, dissol ved Hg 0 can be oxi di zed and la ter com ple xed with ma jor inor ga nic li gands (Cl -, OH -, S 2-, etc.) and dis sol ved or ga nic car bon 4,7,8 , thus in cre a sing the me tal so lu bi lity. So lar ra dia ti on is anot her im por tant va ri a ble in the mer cury aqua tic che mistry, but this has scar cely been ex plo red in the re cent ye ars, es pe ci ally for tro pi cal are as 7,9,10 . Other im por tant pro ces ses re la ted to the to xi city of mer cury, not only to the bi o ta, but ma inly to hu mans, is meth yla ti on by enz yma tic ac ti on of mi cro or ga nisms 4,11,12 , fa vo ring the ac cu mu la ti on of or ga nic mer cu ri al spe ci es in fatty tis su es.
Des pi te the vast li te ra tu re con cer ning both the to xi city and the che mistry of mer cury in aque ous me dia, stu di es are cen te red on Hg 2+ and or ga nic mer cu ri al spe ci es only. Sin ce in the ga rim pos of Bra zil mer cury is al most so lely used in the ele men tal form, very lit tle is known about the fate of this spe ci es in tro pi cal wa ters, spe ci fi cally con cer ning the flux at air/wa ter as well as, wa ter/se di ment in ter fa ces, thus ma king dif fi cult the risk as sess ment, as ex tra po la ti on from pu blished data to the Bra zi li an sce ne has very li mi ted va li dity.
In this work some fun da men tal as pects of the aqua tic che mistry of ele men tal mer cury was stu di ed. The se as pects in vol ved the ad sorp ti on of both Hg 2+ and Hg 0 onto par ti cula te ma te ri al in sus pen si on, as well as the re ac ti ve dis so lu tion of Hg 0 in na tu ral wa ters, fol lo wed by the ex tent of oxi da ti on to Hg 2+ with con co mi tant in cre a se in the con centra ti on of so lu ble mer cury and the los ses of Hg 0 to the atmosp he re. The ex pe ri men tal con di ti ons were set in a man ner that the re sults could be used as an ade qua te tool to sup port many on go ing stu di es on the im pact of mer cury in the se re gi ons.

Expe ri men tal
The Envi ron men tal Che mistry La bo ra tory, whe re this re se arch was car ri ed out (Insti tu te of Che mistry, Sta te Univer sity of Cam pi nas) works un der Good La bo ra tory Prac tices (GLP). All glass wa re, in clu in ding re ac ti on ves sels, was so a ked in 10% ni tric acid pri or to use, fol lo wed by rin sing with dis til led wa ter.

Mer cury de ter mi na ti on
Mer cury was de ter mi ned using a com mer ci ally ava i lable Cold Va por Ato mic Absorp ti on (CVAAS) equip ment adap ted to FIA (Flow Injec ti on Analy sis) as des cri bed by Pas qui ni et al. 13 . Bri efly, the system con sists of the in jec tion of the aque ous sam ple in a car ri er con ta i ning a 5% so luti on (v/v) of HNO 3. The mix tu re re ce i ves the re duc tant (a 10% (w/v) SnCl 2 so lu ti on) to pro mo te the re duc ti on of Hg 2+ ions to the ele men tal form, which is then pur ged from the so lu ti on in a gas-liquid se pa ra ti on cham ber. The me tal va por is car ri ed to the de tec ti on cell af ter an op ti o nal (not used in this case) pre-concentration step con sis ting of a gold amal ga ma ti on co lumn. Dif fe ren ti a ti on bet we en to tal (re du ci ble) mer cury and Hg 0 spe ci es was ob ta i ned when the sam ple was analy zed with and wit hout using the re ductant. The con cen tra ti on of the Hg 2+ spe ci es was ob ta i ned by dif fe ren ce. The li mit of de tec ti on ob ta i ned using this method was 0.1 µg L -1 of mer cury at an analy ti cal fre quency of 50 sam ples h -1 . De ta i led in for ma ti on about the who le analy ti cal pro ce du re is pro vi ded el sew he re 13 .

Re ac ti ve dis so lu ti on of Hg 0 in na tu ral wa ters
Dis so lu ti on of ele men tal mer cury was car ri ed out in sea wa ter (col lec ted at Uba tu ba -SP, Bra zil) as well as in fresh wa ter (UNICAMP lake, Cam pi nas -SP, Bra zil) using si mi lar se tups as des cri bed be low. Expe ri ments using distil led wa ter were also per for med for com pa ri son. So lu ti ons were con ta mi na ted with 1 g L -1 of Hg 0 and the con cen tra tion Hg 0 and Hg 2+ was mo ni to red at pre-established time inter vals. Du ring the who le length of the ex pe ri ment, so lu ti ons were mag ne ti cally stir red at a rate of ap pro xi mately 120 rpm. A si mi lar ex pe ri ment was done using a 0.1 mol L -1 so lu ti on of NaCl. Pri or to the ex pe ri ment, the fol lo wing wa ter pa ra me ters were mo ni to red: tem pe ra tu re, pH, chlo ri de con cen tra ti on, par ti cu la te ma te ri al and TOC (to tal or ga nic car bon), and all ex pe ri ments using na tu ral wa ters were per for med using un fil te red sam ples. La ter on, ex pe ri ments using 10 and 100 g L -1 of me tal lic mer cury and dis til led wa ter were done for com pa ri son. Fi nally, the expe ri men tal set up and con di ti ons iden ti cal to the ones used by Glew and Ha mes 14 to de ter mi ne the so lu bi lity of Hg 0 were re pro du ced using 100 g L -1 of ele men tal mer cury.

Adsorp ti on of Hg 2+ and Hg 0 onto the par ti cu la te sus pen ded ma te ri al
Par ti cu la te ma te ri al was ob ta i ned from a se di ment collec ted at the UNICAMP Lake. Lar ger par ti cles in the wet se di ment were se pa ra ted by using a 1 mm Nylon si e ve. After drying at room tem pe ra tu re, the se di ment was furt her si e ved in to 200 and 400 mesh fra ti ons. Adsorp ti on ex pe riments were car ri ed out at room tem pe ra tu re (23  3 °C) using 2 L be a kers un der cons tant mag ne tic stir ring, by adding 150 or 15 mg of par ti cu la te in 1500 mL to give a fi nal con cen tra ti on of 100 or 10 mg par ti cu la te L -1 . Con trol was car ri ed out in the ab sen ce of par ti cu la tes. Sus pen si ons were con ta mi na ted with eit her 10 µg L -1 of Hg 2+ or 1 g L -1 of Hg 0 . The se ex pe ri ments were car ri ed out using 200 and 400 mesh se di ments as well as the non si e ved ( in na tu ra ) se diment. The to tal con cen tra ti on of so lu ble mer cury was ex peri men tally de ter mi ned in sam ples with drawn at dif fe rent time in ter vals and the pro cess was la ter mat he ma ti cally mo de led.

Re ac ti ve dis so lu ti on of Hg 0 as a func ti on of pH
The dis so lu ti on of Hg 0 at dif fe rent pH va lu es (4.0, 7.0 and 9.0) was eva lu a ted in the pre sen ce of 0.1 M NaNO 3. The pH was set by using eit her NaOH or HNO 3 so lu ti ons at the be gi ning of the ex pe ri ments.

Re sults and Dis cus si on
The first at tempt to es ti ma te the so lu bi lity of Hg 0 in water was done by Bo nho el fer and Re i chardt 15 , whe re the authors sho wed the exis ten ce of the spe ci es Hg 0 (aq) using the re so nan ce line of mer cury at 253.7 nm. Sin ce then, the so lubi lity of ele men tal mer cury in wa ter has been stu di ed by many ot her re se ar chers and, ex cept for the fact that the concen tra ti on of Hg 0 has va ri ed from 100 g L -1 14,16,17 down to 22 g L -1 18 , the com mon fe a tu re in all the se stu di es is the absen ce of dis sol ved oxy gen and pre sen ce of a re du cing agent to avo id the for ma ti on of the Hg 2+ spe ci es in a clo sed system. As al re ady po in ted out, the se ex pe ri men tal con di tions, alt hough ne ces sary for the pur po se, can not be ex tra pola ted to na tu ral wa ter con di ti ons. One of the fe wer pa pers that de als with the so lu bi lity of ele men tal mer cury using ex pe ri men tal con di ti ons more su i ta ble to the ones found in na tu ral wa ters was re cently pu blis hed by Ma ga lhães and Tu bi no 17 , whe re they used air and CO 2 in the aque ous phase, but very high ini ti al con cen tra ti ons of ele men tal mercury.
The fate of mer cury in na tu ral wa ters in vol ves three ma jor steps, two of them oc cur ring in the bulk so lu ti on and the ot her one at the wa ter/at mosp he re in ter fa ce. The first step is the so lu bi li za ti on of mer cury from the li quid pha se to the aque ous pha se, as in Eq. 1. The se cond step is the possi ble oxi da ti on of the dis sol ved spe ci es, as pre sen ted in Eq. 2, de pen ding on the re dox con di ti ons of the wa ter body. The pos si bi lity of the di rect oxi da ti on of mer cury dro plets pre sent in the wa ter can not be ru led out, alt hough it should be very li mi ted sin ce it is an area-dependent pro cess. Finally, Hg 0 spe ci es pre sent in the aque ous pha se will be at equi li bri um with the overl ying at mosp he re, de pen ding on the par ti al pres su re of this spe ci es in the gas pha se, ac cording to Henrys law 2 .
In the oxi da ti on of ele men tal mer cury, the re is a consen sus in the li te ra tu re that Hg 2 2+ are short-lived spe ci es in most na tu ral wa ters, and the ir con cen tra ti on should be ne gli gi ble. Ho we ver, with the re cent de ve lop ment in mercury spe ci a ti on tech ni ques, this sta te ment ne eds furt her study.

Re ac ti ve dis so lu ti on of Hg 0 in na tu ral wa ters
Ele men tal mer cury has been con si de red to be very unre ac ti ve in wa ter bo di es due to the low so lu bi lity of Hg 0 . So lu bi lity of mer cury ob ta i ned in la bo ra tory stu di es shows va lu es around 60 µg L -1 at 25 °C 14,16,18,19 . Inde ed, when trying to de ter mi ne the so lu bi lity of ele men tal mercury in dis til led wa ter, using the same con di ti ons as menti o ned in the li te ra tu re 14 , we ob ta i ned 61 µg L -1 . Ho we ver, in most im pac ted wa ters, even in the hot spots near the garim pos (Ma de i ra Ri ver, Bra zil), the hig hest con cen tra ti on of to tal dis sol ved mer cury found was 0.44 µg L -1 20 , which in di ca tes that eit her na tu ral wa ters are far from the equi libri um, or that ot her pro cess are also im por tant in mer cury so lu bi li za ti on.
Fi gu re 1 shows the in cre a se in the so lu ble con cen tra ti on of mer cury (Hg 2+ and Hg 0 ) as a func ti on of time in a freshwa ter sam ple (UNICAMP Lake). The re sults ob ta i ned using dis til led wa ter are also shown for com pa ri son, and in both ca ses the ini ti al con cen tra ti on of Hg 0 was 1 g L -1 . When using dis til led wa ter, the con cen tra ti on of so lu ble mer cury in cre a ses as a func ti on of time, re a ching the ste ady sta te af ter ~24 h at 5.4 µg L -1 . Inte res tingly, the con cen tra tion of the star ting mer cury spe ci es (Hg 0 ) re a ches the ste ady sta te in a shor ter time, around 5 h, at a con cen tra ti on of 1.7 µg L -1 . For the lake wa ter sam ple, so lu ble mer cury con centra ti ons are lo wer than the ones ob ser ved for dis til led wa ter, but the same trend was ob ser ved: whi le the ele men tal mercury re a ched 0.6 µg L -1 , so lu ble mer cury was clo se to 3.0 µg L -1 .
In ge ne ral, ste ady sta te con cen tra ti ons for both Hg 2+ and Hg 0 spe ci es in the lake wa ter sam ple were about half of the ones ob ta i ned for dis til led wa ter, in di ca ting the pre sence of me cha nisms that inhi bit ele men tal mer cury dis so lu tion. Con si de ring that this na tu ral sam ple had pH = 7.1, TOC = 5.8 mg C L -1 , and To tal Sus pen ded So lids (TSS) = 25 mg L -1 , this inhi bi tory ef fect was at tri bu ted to the par ti cu la te ma te ri al, as will be shown lat ter. Also, another im por tant as pect ob ser ved in the se ex pe ri ments was the oxi da ti on of Hg 0 in both ca ses, a pro cess that oc cur red to a lar ge ex tent sin ce the ra tio Hg 2+ /Hg 0 va ri ed bet we en 3 and 5 for dis til led wa ter and la ke wa ter sam ples res pec ti vely. This oxi da ti on, very li kely due to dis sol ved oxy gen (sin ce the re are no ot her oxi di zing agents in dis til led wa ter) shows that ele men tal mer cury is not inert in oxy gen-rich wa ters, thus in cre a sing the so lu bi lity of the me tal.
When this same ex pe ri ment was car ri ed out using elemen tal mer cury at con cen tra ti ons of 10 and 100 g L -1 in distil led wa ter and in the ab sen ce of re du cing agents (not shown), the to tal so lu ble con cen tra ti on in the ste ady sta te was 15.9 and 316 µg L -1 res pec ti vely, re in for cing the idea that sur fa ce area at the in ter fa ce li quid mer cury/so lu ti on is a very im por tant pa ra me ter con trol ling the re ac ti ve dis solu ti on pro cess.
Re ac ti ve dis so lu ti on of ele men tal mer cury was also inves ti ga ted in se a wa ter to eva lu a te the pos si ble ef fects of sali nity (and chlo ri de com ple xa ti on) in com pa ri son to fresh wa ter sam ples. Sa ne ma sa 21 sho wed that the dis so lu tion of the me tal lic mer cury in elec troly tic so lu ti on de pends on both the con cen tra ti on and the na tu re of the ani on present in so lu ti on. In ge ne ral, the io nic strength ef fect on disso lu ti on of a non-electrolyte is the re sult of phe no me na like hydra ti on, van der Wa als in te rac ti ons and in ter nal pres sure 16,21,22 .
In ex pe ri ments using se a wa ter (pH = 8.12; TOC = 2.77 mg L -1 ; sa li nity 33.10 % o and 13 mg L -1 of TSS) a steady in cre a se in the con cen tra ti on of so lu ble mer cury was ob ser ved, re a ching a peak of 17.8 µg L -1 af ter 10 h. After this peak, the so lu ble con cen tra ti on drop ped, re a ching the ste ady sta te va lue of 4.8 µg L -1 (Fig. 2). Inte res tingly, this same be ha vi or was ob ser ved when the ex pe ri ment was carri ed out using a 0.1 M so lu ti on of NaCl ins te ad of se a wa ter, ex cept for the fact that the fi nal con cen tra ti on of so lu ble mer cury spe ci es were hig her, in this case, than the ones obser ved for se a wa ter. In both ca ses, the tran si ent peak in the con cen tra ti on of so lu ble mer cury was due to the mer cu ric spe ci es only. A pos si ble ex pla na ti on for this is the for ma tion of sta ble chlo ri de com ple xes (HgCl 4 2-, HgCl 3 -, HgCl + , HgCl 2 , etc.), as al re ady shown to oc cur 17 . No net he less, the in cre a se, fol lo wed by a de cre a se in the con cen tra ti on of Hg 2+ spe ci es in the se wa ters is yet to be ex pla i ned.
Anot her im por tant pa ra me ter re la ted to dis so lu ti on in the aqua tic che mistry of mer cury is va ri a ti on of pH. Cheng Yan et al. 4 stu di ed the con cen tra ti on of to tal mer cury and methylmer cury in the Adi ron dack la kes (nort hern New York Sta te-USA), which are cha rac te ri zed by aci dic wa ter, low io nic strength and high con cen tra ti on of dis sol ved orga nic car bon. The re sults ob ta i ned show that the re is an in -cre a se in or ga no mer cu ric spe ci es with a de cre a se in pH. In the pre sent work, the re ac ti ve dis so lu ti on of me tal lic mercury was car ri ed out at three dif fe rent pH va lu es (4.0, 7.0 and 9.0). As shown in Fig. 3, the to tal so lu ble con cen tra ti on of mer cury in the ste ady sta te is very si mi lar for pH va lu es of 7.0 and 9.0, re a ching va lu es around 3 µg L -1 . In this case, the con cen tra ti on of the spe ci es Hg 0 that con tri bu tes to the pool of all aque ous mer cury ac counts for 74% and 58%, res pec ti vely. Ho we ver, when the same ex pe ri ment is car ried out at pH 4.0, the so lu ble con cen tra ti on is much hig her, re a ching 20 µg L -1 . From this fi gu re, only 18% (3.6 µg L -1 ) re pre sents the Hg 0 spe ci es. The pos si ble ex pla na ti on for this be ha vi or is the oxi da ti on of me tal lic mer cury in aci dic pH, ca u sing an in cre a se in the so lu ble con cen tra ti on of the me tal. Accor ding to Amyot et al. 7 , this is ex pec ted in na tural wa ters con ta i ning mi nu te amounts of hydro gen pe ro xide, which act as an oxi di zing agent in aci dic pH, whe re as un der al ka li ne con di ti ons the re duc ti on of mer cu ric spe ci es is the do mi nant me cha nism.
This ob ser ved trend is very im por tant when one ta kes into con si de ra ti on that in the Ama zon re gi on, the oc cur rence of ri vers sho wing very aci dic wa ters is com mon. In the Rio Ne gro ba sin, for ins tan ce, the ave ra ge pH is 4.3 for the so-called black wa ter ri vers, which are very rich in dis solved or ga nic car bon and poor in par ti cu la te ma te ri al, con diti ons con si de red to be very su i ta ble for in cre a sing the eco to xi co logy of mer cury 5 .
Adsorp ti on of Hg 0 and Hg 2+ spe ci es onto the sus pen ded ma te ri al The role of par ti cu la te ma te ri al in the re ac ti ve dis so lu tion of Hg 0 as well as in the ad sorp ti on of so lu ble Hg 2+ spe cies were also ad dres sed in this study. This is a very im por tant pa ra me ter in or der to mo del to mi mic the real sce na rio, sin ce the ga rim po ac ti vi ti es also con tri bu te to a mas si ve in cre a se in the amount of sus pen ded so lids in the iga ra pés (small ri vers). This ad di ti o nal im pact as so ci a ted with the ga rim pos has been con si de red as be ing more cri ti -cal on a short-time ba sis to the ecosy stem than the use of mer cury it self.
The ef fects of the par ti cu la te ma te ri al (100 mg L -1 of a 400 mesh ma te ri al) on the re ac ti ve dis so lu ti on of me tal lic mer cury is shown in Fig. 4. When in the pre sen ce of par ticu la te ma te ri al, the so lu ble con cen tra ti on of mer cury re aches about 1 µg L -1 , whe re as in the ab sen ce of ad sor bent sur fa ces, the con cen tra ti on is three ti mes hig her. The se results are si mi lar to the ones shown in Fig. 1, ex cept for the fact that in this case the amount of par ti cu la te ma te ri al is four ti mes hig her, but the re is no or ga nic ma te ri al such as pre sent in the lake wa ter sam ple. Although the inhi bi tory ef fect of the par ti cu la te ma te ri al on the re ac ti ve dis so lu ti on of mer cury was de ter mi ned, it was not pos si ble to de tect which spe ci es (Hg 0 or Hg 2+ ) shows the hig her af fi nity for the bin ding si tes on the par ti cu la te ma te ri al.
Fi gu re 5 shows the af fi nity of mer cu ric spe ci es for diffe rent si zes of the par ti cu la te ma te ri al. As ex pec ted, ad -  sorp ti on of Hg 2+ ions fol lows the or der of pre fe ren ce: 400 mesh > 200 mesh > se di ment in na tu ra , whe re the half-life va lu es for mer cu ric ions in so lu ti on is ap pro xi ma tely 5, 15 and 130 mi nu tes res pec ti vely. Mo de ling the di sap pe a ran ce of Hg 2+ spe ci es in the aque ous pha se due to ad sorp ti on onto the so lid pha se was car ri ed out using the fol lo wing mat he ma ti cal equa ti on: dC dt a bC = + whe re C is Hg 2+ con cen tra ti on at a gi ven time, and a and b are cons tants. In this par ti cu lar case, Eq. 1 can be sol ved after in te gra ti on: whe re a/b is the mer cury con cen tra ti on in the aque ous phase in equi li bri um with the so lid pha se. Va lu es of a and b, ob ta i ned un der the ex pe ri men tal con di ti ons de ta i led in Fig.  5, are shown in Ta ble 1.

Con clu si ons
The use of mer cury by the ga rim pos spre ad throug hout the Ama zon re gi on has al re ady been con si de red as a time bomb 23 , and very of ten com pa red to a new Mi na ma ta. Howe ver, this com pa ri son stands wit hin li mi ted sci en ti fic infor ma ti on, es pe ci ally when one ta kes into con si de ra ti on the fact that the mer cury be ing re le a sed in the Ama zon is mostly me tal lic mer cury, and not in or ga nic forms as in Mina ma ta.
Con si de ring that the li te ra tu re about the fate of ele mental mer cury in the aqua tic system (and more spe ci fi cally tro pi cal wa ters) is no ne xis tent, eva lu a ting the real im pact of the re le a se of mer cury into this very uni que ecosy stem is still dif fi cult. This pa per brings some new in sights about the fate of Hg 0 in the aque ous system, in di ca ting that the situ a ti on fo re cas ted for the Ama zon may not be as pes si mistic (in a short time) as the one of ten ex plo red. The se re sults also show that me tal lic mer cury is re ac ti ve in na tu ral waters, un der go ing re ac ti ve (oxi da ti ve) dis so lu ti on. This process, which al ters the spe ci a ti on of mer cu ric forms, le ads to the pro duc ti on of a more to xic form, more su i ta ble to methyla ti on and a more mo bi le spe ci es than the one pre vi ously con si de red.