Solvated Electron

to 10¯6. M is reported. These compounds react with e¯aq with rate constants that vary from 4 χ 109 M¯1 ... saturated 0.0012V NaOH solution rendere...
1 downloads 0 Views 1MB Size
18 Submicromolar Analysis of Hydrated Electron Scavengers EDWIN J. HART and E. M. FIELDEN

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

Department of Chemistry, Argonne National Laboratory, Argonne, Ill.

The analysis of typical e¯ scavengers such as aq

oxygen, hydrogen

peroxide, nitrous

oxide,

acetone, and thymine in the concentration range from 10

¯8

to 10

¯6

M is reported.

compounds react with e¯

aq

These

with rate constants

that vary from 4 χ 1 0 M ¯ 1 s e c . ¯1 for thymine to 9

1.9 χ

10

10

M

¯1

sec.

¯1

for oxygen.

electrons are introduced

into an

Hydrated aqueous,

hydrogen-saturated matrix at pH 11 by 2 rad x-ray pulses of 0.4 to 2 μ sec. duration.

From

the half-life of e¯ , obtained by absorption spec­ aq

trophotometry, the concentration of the dis­ solved species may be estimated.

Some re­

sults are reported on the effectiveness of stand­ ard evacuation and irradiation techniques for the removal of oxygen from aqueous solutions. The mechanisms of the e¯

aq

H and N O + H 2

2

2

initiated

HO 2

2

+

chain reactions are briefly

discussed.

p u r i n g the course of recent work (8), it became evident that the h y ­ drated electron, (e~ ), possesses many of the characteristics of an ideal analytical reagent. It is readily generated, highly reactive, deeply col­ ored, and capable of rapid and precise determination. B y utilizing these properties, it is possible to analyze a wide variety of inorganic and organic compounds at submicromolar concentrations with 5 to 10 X 10 ~ Mel formed b y pulsed x-rays. T h i s paper describes the tech­ niques used to analyze oxygen, hydrogen peroxide, nitrous oxide, acetone, and thymine i n an aqueous matrix at p H 11. aq

9

q

Basis of

Method

Hydrated electrons, generated by short pulses of x-rays or electrons absorbed i n water, react with many molecules and ions with high rate 253

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

SOLVATED ELECTRON

254

constants (3, 6, 9, 10). Consequently, i n the presence of such e~ scavengers, decay of e~ is rapid. O n the other hand, especially i n hydrogen-saturated alkaline solutions where H is neutralized and O H is rapidly converted to H and then to e~ (see Reactions 4 and 5 below), the decay of e~ is comparatively slow at concentrations < 10 ~ M. T h e decay reaction is aq

aq

+

ag

8

aq

+ elq - * H

2

+ 20H-

(1)

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

whereas i n the presence of an excess of a n electron scavenger—i.E., O 2 , the decay is predominantly e'

+ 0

aq

(2)

0 ~

2

2

A t a n [el ] of 10 ~ M, the theoretical n / owing to Reaction 1 alone is 10 msec, compared to 0.35 msec, i n the presence of 10 ~ M 0 . Since ab­ sorption spectrophotometry b y photoelectric recording techniques (4, 6, 11) will readily follow half-lives of the order of 10 ~ sec., measurements under these conditions are routine. 8

g

2

7

2

6

Experimental P r e p a r a t i o n o f M a t r i x . O u r standard matrix consists of a n H saturated 0.0012V N a O H solution rendered free of 0 and e~ scavengers by degassing and irradiation techniques (8,14). About 700 m l . of triply distilled water is given a preliminary degassing i n a one liter evacuation chamber i n order to remove most of the 0 and C 0 . T h e n 1.02V N a O H is added to make the solution 0.0012V. T h e solution need not be carbon­ ate free, but it is advisable to minimize this ion because it forms a n absorb­ ing transient ion b y reacting with the O H radical (1). Next, the solu­ tion is saturated with H , degassed (cycle repeated twice), and finally satu­ rated with H . T h i s solution is then forced into H purged 100 (or 50) m l . syringes. T h e residual 0 and e~ scavengers are removed b y 15 min. C o 7-ray irradiation to a total dose of ~ 1 0 0 μΜe~ (8). P r e p a r a t i o n o f S o l u t i o n s . Standard solutions of 0 , H 0 , N 0 , thymine, and acetone are prepared either i n the pre-irradiated matrix or pre-irradiated H saturated water at concentrations ^ 1 0 to 50 μΜ. A l i quots of these solutions are then micropipetted into the irradiation cell containing the matrix (Figure 1). T h e microsyringe is filled from a larger syringe b y forcing the solution to be analyzed into the needle, up through the microsyringe barrel, and out of the top. After flowing freely for some seconds, the plunger is reinserted. T h i s technique thoroughly eliminates air from the syringe, a serious contaminant because of its oxygen. I r r a d i a t i o n C e l l A s s e m b l y . Our 13.2 m l . irradiation cell (2.2 c m . diameter, 4 c m . length), provided with a glass encased magnet, is shown in Figure 1. T h i s cell is mounted i n a multiple reflection assembly (13). I n normal practice this cell is used with 16 light passes giving a n optical path length of 64 c m . T h e matrix, stored i n a 100 m l . syringe, is intro­ duced into the He-purged cell through stopcock A . After two complete replacements, 5/20 cone Β on the H e line is replaced b y microsyringe C containing the sample. Because traces of oxygen and other impurities may have been introduced with the solution, it is pulse irradiated until a 2

aq

2

2

2

2

2

2

2

e 0

aq

aq

2

2

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

2

2

2

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

18.

HART AND FIELDEN

Figure

1.

255

Submicromolar Analysis

Irradiation cell assembly for e ~ spectrophotometry. aq

See reference (7) for optical arrangement and reference (13) for multiple reflection cell details. constant decay of e~ results (T I ~ 8-10 m s e c ) . T h i s treatment usually requires about twenty 7 n M * e' pulses. ( n M signifies nanomolar (10~~ M ) ) . After this final irradiation, the sample is injected and the solution mixed by agitating the stirrer several times with a magnet. T h e solution is now homogeneous and ready for the x-ray or electron-analyzing pulse. Figure 2 shows the decay of a typical solution before and after injecting 13 n M 0 . Because low concentrations of oxygen (^100 n M ) are analyzed many samples may be r u n without changing the matrix. Irradiation with 100-500 pulses usually restores the solution to its original TM because the final reduction product is water. Other scavengers, too, can often be removed b y repeated pulsing so that the matrix may be reused. R e c o r d i n g o f e~ D e c a y . T h e intensity of the light passing through the reaction cell is monitored b y a conventional, monochromator/photoaq

L

2

aq

9

2

aq

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

SOLVATED ELECTRON

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

256

multiplier/oscilloscope combination. T h e photomultiplier was a n R C A 7102 with associated pre-amplifier. T h e rise time of the pre-amplifier could be controlled by switching i n various integrating capacitors. T h i s has the effect of removing the high frequency components of the ran­ dom noise fluctuations present i n the signal. Care was always taken to ensure that the band width of the electronic circuitry was adequate to dis­ play the transient absorption being recorded. I n practice the rise time of the recording system was never greater than 5 % of the half-life of the transient signal. F o r example i n the oscilloscope traces of Figure 2 the rise time (sometimes called response time or integrating time) of the com­ plete recording system was 100 μββα and the half-lives of the two decays are 8000 and 2000 #sec. Reducing the rise time to 10 jttsee. did not affect the experimental half-lives showing that negligible distortion had been produced by the 100 μββα rise time.

Figure 2. Decay of eâq absorption band at 6900 A. in a hydrogen-saturated solution at pH 11 (a) 0 -free (b) 1.3 X 10- MO . Decay is from right to left. The lower horizontal trace was recorded 30 msec, before the x-ray pulse and represents 100% transmission. 2

8

2

T h e standard procedure for taking a picture like Figure 2 is to adjust the photomultiplier pre-amplifier output to be 2 volts—i.E., the difference between 100% and zero light is represented by this voltage. T h e signal is then amplified b y a known factor (50 i n the case of Figure 2), and the radiation pulse is triggered. T h e upper line i n Figure 2 represents the 100% light-level and is recorded 30 msec, before the radiation pulse occurs. T h e two lower traces are superposed examples of transient absorptions produced b y the pulse. I r r a d i a t i o n S o u r c e . One to two rad electron or x-ray pulses are required to produce 6 to 12 n M e' . W e use 1 jusee. pulses of 16 M.E.v. tungsten x-rays generated with an A R C O electron accelerator. T h e pulse must, however, be introduced in a time short compared to the measured half-lives. A n y similarly pulsed x-ray beam of 150 to 200 k.E.v. would serve equally well since there is no rigid requirement of uniform irradiaaq

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

18.

HART AND

FIELDEN

257

Submicromolar Analysis

tion. W i t h x-ray pulses of lower energy, more pulses may be required i n order to clean out the scavenger between runs. Alternatively, a steady x-ray or y-ray source may be substituted for the pulsed x-rays for this purpose. C a l c u l a t i o n o f R e s u l t s . B y considering the loss of e~ as two psuedo—first-order reactions, namely, a decay b y the matrix, m , and decay b y the scavenger, s aq

+ m -> m~"

eâq e'aq

+ s -> s~

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

we have where k

m

= Me«-«)(m) +

and k, are rate constants for matrix and solute, respectively. ln(e; ) = [* (m) + e

for

= y

(elg)

k,(elq)(B)

2

and since

(e )o aq

s

w

k.(s)]t

= 0.693/n/

k (m) m

(1

=

k,

2

L\

\til2

Til 2/

k, is the rate constant for the scavenger, s, tn is the half-life of the solute reaction, and r i / is the half-life of the matrix. T h e matrix reaction only approximates first-order since Reaction 1 i n an absolutely pure matrix is second-order. In general, trace impurities improve the first-order approximation. Since the matrix reaction only accounts for a small fraction of the decay i n the presence of scavengers, the errors involved i n this approximation are negligible compared to other experimental errors. 2

2

Results Analysis of el scavengers at concentrations i n the range from 10 ~ to 10 ~* M has been demonstrated for oxygen, hydrogen peroxide, nitrous oxide, acetone, and thymine at p H 11. In principle, any inorganic or organic compound reacting with e~ with rate constants > 10 M sec." may be analyzed with similar results. F o r compounds with lower rate constants the sensitivity of the method diminishes b y a factor of 10 for each order of magnitude decrease i n rate constant. O x y g e n . T h e results of several runs carried out at 0 concentrations from 18-190 n M are shown i n Figure 3. While there is some scatter in the points, considering the extremely low concentrations dealt with and the difficulties i n handling gas solutions, this calibration curve provides a satisfactory demonstration of the feasibility of eâq analyses. T h e ordinate i n Figure 3 represents the calculated oxygen concentrations i n the irradiation cell based on the experimental e~ decay half-life and the value for kr + 0 of 1.9 X 10 M~ sec." (7). T h e points should lie on the indicated line if the above rate constant is correct under these conditions. 8

Q

9

aq

-

1

1

2

aq

aq

2

10

l

1

T o illustrate a practical use of our method, we have measured the "0 -equivalent concentrations" of some matrix solutions after different 2

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

SOLVATED ELECTRON

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

258

20

40

60

80

100

120

140

160

180 200 220

added oxygen concentration (10 ~ M) β

Figure

3.

Determination of oxygen from eiq half-life measurements.

Δ pH 11, Ht-saturated matrix Θ pH 11, 0.001 M CHzOH matrix.

treatments. F o r example, how effective are our evacuation and irradia­ tion techniques for removing 0 ? H o w much 0 is introduced into the irradiation cell b y a typical transfer of matrix from a 100 m l . syringe into a He-purged cell? F o r the first time, we can place an upper limit on the oxygen content of aqueous solutions treated in these ways. T h e effect of a number of evacuations on oxygen removal is shown i n experiments 1-3 of Table I. Calculation of the oxygen concentration is based on the assumption that 0 is the only e~ scavenger present. T h e results show that little is to be gained b y more than one evacuation and subsequent hydrogen saturation. 7-ray irradiation is a n extremely effective method of removing 0 i n our H -saturated matrix. See Experiments 4 and 5 of Table I. Irradia­ tion with about 25 μΜ ei reduces the 0 concentration to < 0.03 μΜ 0 . Further irradiation does not lower the 0 level. T h i s result indicates that some air or impurities leak into the syringe. Pulse irradiation is the most effective way of removing the final traces of oxygen from the reaction cell. Experiments 6a-6g display the progress of 8 n M e~aq x-ray pulses in removing oxygen from sample 6 that had previously been C o irradiated with 75 μΜ e~ . Experiments 7-7d show the disappearance of oxygen after 69 n M had been injected into matrix 2

2

aq

2

2

2

q

2

2

2

60

aq

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

18.

HART AND FIELDEN

259

Submicromolar Analysis

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

Table I.

Exp. No. 1 2 3 4 5 6 6a 6b 6c 6d 6e 6f 6g 7 7a 7b 7c 7d

(ft) nM 2500 2500 1200 30 41 69 69 48 28 11 6 2 0 69 60 46 26 4

Oxygen Measurements in pH 11 Matrix Pretreatment No. No. of of Ά μΜ(β$ Evacu­ équilibra- (Co ' Δ(0 )« te) te)/ nM A(ft) nM ations Hons rays] 1 1 0 0 2 2 0 3 3 0 3 3 25 0 3 3 50 0 3 3 75 16 20 80 4.0 40 160 4.0 58 240 4.1 63 320 5.1 66 400 6.1 69 800 11.6 0 8 9 40 4.5 23 80 3.5 43 200 4.7 65 600 9.2 80

6

2

* Decrease in Os concentration during pulse irradiation in Experiments 6 and 7. Accumulated dose of (eaq) introduced by x-ray pulses. b

6g. Note i n column 5 that each of these groups of experiments requires 4 eâç/molecule 0 i n the initial stages of 0 removal. T h i s is the theoretical number of eâq needed to convert 0 into 20 ~ . O t h e r elq Scavengers. Calibration curves i n Figure 4 for hydrogen peroxide, nitrous oxide, acetone, and thymine illustrate the precision and versatility of the method. A s with 0 , H 0 may be satisfactorily analyzed at concentrations below 100 n M . Acetone, nitrous oxide, and thymine with somewhat lower eâq rate constants are less sensitive, but 5-10% precision is possible for solutions of 500 n M concentrations. T h e ordinate i n Figure 4 is the experimental quantity (1/fe/a — l / r i / ) , and the slopes of the calibration curves are a measure of the rate constants of the e'aq scavenger reactions. Table II lists rate constants obtained from the data i n Figures 3 and 4 compared to previous literature values. Estimation of nitrous oxide and hydrogen peroxide presents a special problem i n that a product of their reactions with hydrated electrons is an O H radical. One of the features of the hydrogen-saturated p H 11 matrix is the conversion of O H radicals into eâ , so that with these solutes decomposition will occur by a chain reaction—i.E., for H 0 : 2

2

2

2

2

2

2

2

q

2

eâq + H 0

2

OH + H

2

2

H

2

O H ~" + O H — H 0 + H 2

+ O H - - * H

2

0

+elq

(3) (4) (5)

and for nitrous oxide: eâq + N 0 2

N 0-? 2

N

2

+ OH

followed b y (4) and (5).

In Solvated Electron; Hart, E.; Advances in Chemistry; American Chemical Society: Washington, DC, 1965.

(6)

SOLVATED ELECTRON

Downloaded by UCSF LIB CKM RSCS MGMT on December 3, 2014 | http://pubs.acs.org Publication Date: January 1, 1965 | doi: 10.1021/ba-1965-0050.ch018

260

added scavenger concentation (10"*·)

Figure 4.

Calibration curves for some e~ scavengers at pH 11. aq

tm = half-life with scavenger; n/ = half-life of matrix. Θ H 0 in 0.001 M CH 0H; solid line, + ,o = 1.3 X 10 M~ sec.Φ Acetone in sat. H ; solid line, k*L + acetone = 6.9 X 10 M~ sec. © N2O in 0.001 M CH OH; solid Une, k