Open Access

40Ar/39Ar age determination using ARGUS VI multiple-collector noble gas mass spectrometer: performance and its application to geosciences

Journal of Analytical Science and Technology20156:4

DOI: 10.1186/s40543-015-0049-2

Received: 19 October 2014

Accepted: 19 January 2015

Published: 12 February 2015

Abstract

Background

Ar/39Ar dating technique has been used to determine the age for low-temperature geological event. The introduction of the multiple collectors and the improvement in sensitivity in the noble gas mass spectrometry enable the single-grain 40Ar/39Ar age determination.

Findings

The protocol for 40Ar/39Ar age determination and performance of the new high-sensitivity noble gas mass spectrometer (ARGUS VI) combined with the infra-red laser heating device are shown.

Conclusions

ARGUS VI can produce the precise single-grain 40Ar/39Ar age of 28.4 ± 0.5 Ma (1σ, MSWD = 138, n = 24) within the recommended values of Fisher Canyon sanidine (FCS) (28.294 ± 0.036 Ma). For the young samples, we can also get the age of 1.185 ± 0.004 Ma (1σ, MSWD = 6.05, n = 26) of Alder Creek sanidine (ACS), equivalent to the recommended age (1.193 ± 0.001 Ma).

Keywords

40Ar/39Ar age Noble gas mass spectrometer Quaternary Geochronology CO2 laser

Findings

Introduction

Recent development in mass spectrometry, such as the increase of sensitivity and the adoption of multiple collector system, enables to analyze the isotopic ratio for the single grain and spot analysis. Since the introduction of in-situ age dating for the small geological samples such as SHRIMP (sensitive high-resolution ion micro probe), the complex geological events can be revealed more precisely. The U-Pb spot age dating for zircon grains suggests the new concept of early Earth (e.g., Wilde et al. 2001) and provides the important keys to solve the juxtaposed geological events in a certain region (e.g., Cheong et al. 2014). However, to reveal the full history of a certain orogeny, it is indispensable to get the age information for the low- to mid-temperature geological activity. The Ar isotope system, whose closure temperatures range from ca. 130 to 690°C (e.g., Faure and Mensing 2005), can provide good clue to solve it. In addition, for the basaltic rocks with rare minerals adequate for U-Pb dating, the Ar ages have become the good proxy for the eruption age. For these purpose, K-Ar age dating method was introduced in early 50s and had provided age information related to various studies, such as the geomagnetic polarity timescale, one of the essential foundations of plate tectonic concepts (McDougall, 2014). However, the possible inhomogeneity due to the separate analyses of K and Ar and no information about the Ar-loss and/or Ar-gain after the closure in Ar isotope system should be considered for the proper interpretation of K-Ar ages. The 40Ar/39Ar technique, a variant of K-Ar dating system, has now commonly replaced the conventional K-Ar method to overcome these disadvantages. In addition, the introduction of laser heating/ablation technique enables to measure Ar age of single grains and monitor the variation in Ar isotope within minerals. In this technical note, we describe the 40Ar/39Ar age dating protocol using the newly installed multi-collector noble gas mass spectrometer in Korea Basic Science Institute (KBSI) and also report its performance for some geological samples.

Principle of 40Ar/39Ar age determination

40Ar/39Ar age dating is based on the same principle of K-Ar method, which uses the decay of 40K to 40Ar through the electron capture and positron emission. The concentration of parent 40K is measured indirectly the neutron irradiation process which converts a part of 39 K to 39Ar through the 39 K(n, p)39Ar reaction. The constant 40K/39K ratio in samples enables to estimate the content of 40K in samples through the measurement of 39Ar, if we exactly know how many portions of 39K in samples are converted to 39Ar during the neutron irradiation. The 40Ar/39Ar age is calculated using the following formula (McDougall and Harrison 1999):
$$ \mathrm{t}=\frac{1}{\lambda}\cdot \ln \left(1+J\cdot \frac{{}^{40}A{r}^{*}}{{}^{39}A{r}_{\mathrm{K}}}\right) $$
where
$$ J = \frac{{}^{39}K}{{}^{40}K}\cdot \frac{\lambda }{\lambda_{\mathrm{e}}+{\lambda}_{\mathrm{e}}\hbox{'}}\cdot \varDelta \mathrm{T}\cdot \int \phi \left(\mathrm{E}\right)\sigma \left(\mathrm{E}\right)\mathrm{d}\mathrm{E} $$
t

age

λ

total decay constant of 40K (5.534 × 10−10/year)

λ e

decay constant of 40K to 40Ar by electron capture (0.572 × 10−10/year)

λ e′: 

decay constant of 40K to 40Ar by emission of a positron (0.0088 × 10−10/year)

40Ar*: 

radiogenic 40Ar

39ArK

39Ar derived from 39K by the reaction with fast neutron

ΔT

duration of irradiation

ϕ(E): 

neutron flux density at energy E

σ(E): 

neutron capture cross section at energy E

As the precise values of ϕ(E) and σ(E) are difficult to obtain, standard minerals of age-known are used to calculate J values using the following equation:
$$ J=\frac{ \exp \left(\lambda t\right)-1}{{}^{40}Ar*/{}^{39}A{r}_{\mathrm{K}}} $$

During neutron irradiations, the isotopes of Ca, K, and Cl in the samples commonly produce interfering Ar isotopes through interactions of neutron. Reactor-derived 40Ar from 40K, and 39Ar and 36Ar from 42Ca and 40Ca, respectively, are corrected by analyzing K- and Ca-salts irradiated together with samples. For the detailed calculation for these interfering isotopes, refer to McDougall and Harrison (1999).

Availability and requirements

Outline of the Ar age dating system

40Ar/39Ar age dating system in KBSI can be divided into the following three parts: (1) laser heating system, (2) gas preparation bench, and (3) high-sensitivity noble gas mass spectrometer (Figure 1).
Figure 1

ARGUS VI system in KBSI. (a) Photo of whole 40Ar/39Ar age dating system and (b) schematic illustrations for the gas preparation system.

Laser heating system

Laser heating system (Fusions 10.6, Photon Machines) consists of an integrated CO2 laser source and beam-delivery system with CCD camera and sample chamber. The laser source generates the continuous CO2 laser with the wavelength of 10.5 μm. Gantry type beam delivery system consists of three mirrors, one iris and one focusing lens. Whole laser delivery system also navigates 3-dimensionally over sample chamber. The intensity and beam size of laser including the movement of beam delivery system are controlled by the Chromium II software. The sample chamber is made up with a 114 mm diameter Conflat flange and differentially pumped ZnS viewport by turbomolecular pump. A copper sample holder of 43 mm diameter and 5 mm in thickness with 133 holes is loaded into sample chamber for single grain analysis. KBr glass covers the sample holder to prevent the spillover and scattering of sample from the hole during heating. After sample loading, sample chambers are heated by infrared lamp during several hours to achieve good vacuum level. Released gases from sample by laser beam diffuse from sample chamber into gas preparation system.

Gas preparation system

The outline of the gas preparation system is shown in Figure 1b. All pipe line is made from internally polished stainless steel. In order to purify argon from the extracted gases, three SORB-AC getter pumps (NP10) are used. They are constructed from a cartridge of getter material (ST101 alloy of zirconium with 16% aluminum) placed around an axial heater. At room temperature, these getters pump out hydrogen and carbon monoxide which are major background gases in the mass spectrometer. The getter can be run at 400°C to enhance the pumping of less reactive gases such as hydrocarbons. The vacuum level of gas preparation system reaches approximately 2 × 10−9 mbar by the turbomolecular pump and ion pump. The air of 0.1 cm3 from the automatic pipette system consisting of standard volume and two pneumatic valves is routinely measured to derive the discrimination factor.

Mass spectrometer

The purified argon gas through gas preparation system is directly introduced into the mass spectrometer (ARGUS VI). ARGUS VI is an all-metal single focusing, 13 cm radius, 90° extended geometry magnetic sector mass spectrometer designed for operation in static mode (Mark et al. 2009). A Nier-type electron bombardment source is equipped with y- and z-focusing. All source parameters, including the 5 kV acceleration potential, are computer-controlled and monitored by electronic read-backs. This system is equipped with five Faraday detectors and one compact discrete dynode (CDD) detector for the simultaneous measurement of five isotopes of argon (36Ar, 37Ar, 38Ar, 39Ar, and 40Ar). The 40Ar beam is measured in the high-mass (H1) detector fitted with a 1011 ohm resistor. The 39Ar to 37Ar beams are collected in other Faraday detectors fitted with 1012 ohm resistor. The 36Ar beam is measure by CDD detector. The amplifiers for Faraday detectors are kept in housing evacuated to 1–2 mbar in order to reduce electronic noise. Sensitivity at 200 μA trap current for Ar reaches to 1.15 Amps/Torr and the mass resolution (M/ΔM) is ca. 250. The background value during sample analysis is <1.0 × 10−16 mols 40Ar and <1.8 × 10−18 mols 36Ar in static mode.

System integration and data reduction

Laser heating system, gas preparation system, and mass spectrometer are integrated and controlled by the MassSpec software (Alan Deino Software). It communicates with Chromium II software in laser heating system and Qtegra software in ARGUS VI and controls each parameter in source electronics, magnet control, and steering plate in front of multiple collectors as well as the manipulation of gas preparation system. In addition, it enables the scheduled automatic analysis for tens of analysis. Fitting of raw data, age calculation, and data presentation are also performed using MassSpec software.

Experimental procedures for 40Ar/39Ar age

Neutron irradiation

As suggested in the above section, neutron irradiation process is indispensable for the 40Ar/39Ar dating. For irradiation, several grains of sample were dropped in each hole on aluminum sample disk with 10 holes of 5-mm diameter along with K2SO4 and CaF2 monitors. Age-known standard minerals were placed in the center or margin of sample disk as the neutron flux monitor. Fisher Canyon sanidine (FCS, 28.294 ± 0.036 Ma: Renne et al. 2010) and Alder Creek sanidine (ACS, 1.193 ± 0.001 Ma: Nomade et al. 2005) were used as flux monitor minerals. Two or three sample disks were stacked and wrapped by 0.25-mm-thick cadmium foil in order to reduce the effect of thermal neutrons. Samples were irradiated for 100 h at IP-4(c) position of hydraulic rabbit irradiation facility in HANARO research reactor with the total neutron flux of ca. 3.7 × 1013n/cm2s in Korea Atomic Energy Research Institute (KAERI).

Argon isotope measurement

One or several grains of irradiated samples are placed in a hole on the 133-hole sample holder and then evacuated to ultra high vacuum of ~10−9 mbar. Multiple run sequences for sample analysis including air and blank measurement are executed on the MassSpec software. The heating of samples and subsequent cleaning of extracted gas and the measurement of isotopes are sequentially performed by previously defined procedure file. After heating samples for 6 seconds and purifying gases for another 4 minutes with 3 SAES getters, the argon gas is introduced to the mass spectrometer. Each signal of argon isotopes is measured by the multiple data collection using five detectors using H1 to CDD detector or by the peak jumping method using CDD for 36Ar to 39Ar and H1 detector for 40Ar, respectively. For the baseline, the signals at mass 35.7 and 40.3 are measured at the start and end of the analysis. Blank analyses are made every three unknown sample analysis with the identical condition to the actual samples except for laser heating. For the mass discrimination, standard air is measured after every 10 analysis. Variation of blank and mass discrimination factor during analysis are corrected by parabolic regression. Net intensity of each Ar peaks is extrapolated to the zero time when the argon gas was introduced into the mass spectrometer. After correcting the blank and the mass discrimination, J values and ages are calculated for the standard and unknown samples.

Determination of J value

As neutron flux changes considerably across the sample disk, monitor minerals such as FCS and ACS sanidine are placed in three positions to correct the horizontal gradient in neutron flux. For real samples, the corrected J value along the different position of samples in the sample disk is applied to the age calculation according to the 3-dimensional modeling for the horizontal flux gradient (Figure 2).
Figure 2

3-Dimensional variation in J values across one irradiation disk.

Results and discussion

Single-grain total fusion 40Ar/39Ar analyses were repeatedly performed on two standard materials in order to evaluate the performance of ARGUS VI system (Table 1). Figure 3 shows the age probability diagrams for the total fusion ages from single grains of FCS and ACS. The measured value of FCS (Figure 3a) is 28.4 ± 0.5 Ma (1σ, MSWD = 138, n = 24) which is consistent with recommended value (28.294 ± 0.036 Ma: Renne et al. 2010) within uncertainty. Figure 3b shows the single-grain total fusion ages of ACS sanidine (1.185 ± 0.004 Ma, 1σ, MSWD = 6.05, n = 26), which also coincide with recommended age value (1.193 ± 0.001 Ma, Nomade et al. 2005). Such performance shows that the abovementioned 40Ar/39Ar age protocol can be successfully applied to the Quaternary rocks such as ACS.
Table 1

Representative analysis of 40 Ar/ 39 Ar age for Fisher Canyon sanidine (FCS) and Alder Creek sanidine (ACS)

Relative isotopic abundances

Derived results

Lab ID no.

J

40 Ar

39 Ar

38 Ar

37 Ar

36 Ar

39 Ar Mol

Ca/K

% 40 Ar *

Age (Ma)

(×10 −3 )±1σ

±1σ

±1σ

±1σ

±1σ

±1σ

×10 −14

  

±1σ

FCS

84-02

0.2055

0.0003

56,989

8

748

3

57

3

22

2

1.27

0.09

0.025

0.161

99.4

27.83

0.10

84-03

0.2055

0.0003

96,025

11

1,239

3

87

3

26

2

1.77

0.09

0.041

0.116

99.5

28.32

0.06

84-04

0.2055

0.0003

79,364

11

1,033

7

65

9

9

3

0.56

0.14

0.034

0.050

99.8

28.17

0.20

84-05

0.2055

0.0003

36,927

10

478

7

29

9

7

3

0.14

0.14

0.016

0.087

99.9

28.32

0.42

84-06

0.2055

0.0003

70,660

11

913

7

49

9

10

3

0.87

0.14

0.030

0.065

99.6

28.34

0.22

87-01

0.1959

0.0003

60,413

8

759

3

44

3

14

3

1.99

0.08

0.025

0.100

99.0

27.61

0.10

87-02

0.1959

0.0003

63,977

16

583

3

30

3

0

3

2.47

0.09

0.019

0.000

98.9

33.90

0.18

87-03

0.1959

0.0003

54,528

7

686

3

49

4

11

3

0.83

0.08

0.023

0.091

99.6

27.72

0.14

87-05

0.1959

0.0003

103,304

9

1,264

3

77

2

0

2

3.63

0.09

0.042

0.000

99.0

28.32

0.08

87-06

0.1959

0.0003

46,783

10

581

7

36

8

4

3

0.54

0.13

0.019

0.043

99.7

28.10

0.32

87-07

0.1959

0.0003

35,618

9

445

7

33

8

1

4

0.58

0.13

0.015

0.011

99.5

27.91

0.42

87-09

0.1959

0.0003

78,316

11

969

7

55

8

8

3

1.29

0.14

0.032

0.045

99.5

28.17

0.19

93-01

0.1816

0.0001

63,628

8

724

2

37

3

5

2

1.79

0.09

0.024

0.038

99.2

28.28

0.10

93-02

0.1816

0.0001

127,664

12

1,453

3

70

4

9

3

1.53

0.10

0.048

0.034

99.7

28.41

0.07

93-03

0.1816

0.0001

108,347

10

1,228

3

65

4

7

2

0.59

0.09

0.041

0.034

99.8

28.59

0.08

93-04

0.1816

0.0001

124,168

11

1,424

3

78

4

14

3

2.15

0.09

0.047

0.056

99.5

28.16

0.07

93-05

0.1816

0.0001

51,898

7

595

3

37

3

1

2

1.06

0.08

0.020

0.007

99.4

28.16

0.15

93-06

0.1816

0.0001

100,604

10

1,167

3

77

3

13

3

0.95

0.09

0.039

0.064

99.7

27.90

0.08

93-07

0.1816

0.0001

81,501

9

935

5

52

3

3

3

0.99

0.08

0.031

0.019

99.6

28.18

0.14

93-08

0.1816

0.0001

108,398

9

1,225

5

72

3

51

2

1.04

0.09

0.041

0.233

99.7

28.65

0.11

93-09

0.1816

0.0001

100,317

10

1,154

3

78

3

8

3

3.84

0.09

0.038

0.037

98.9

27.91

0.09

93-10

0.1816

0.0001

120,283

10

1,386

3

96

3

29

2

0.72

0.08

0.046

0.119

99.8

28.13

0.06

93-11

0.1816

0.0001

89,456

9

1,022

3

77

3

25

3

2.80

0.09

0.034

0.139

99.1

28.16

0.08

93-12

0.1816

0.0001

141,018

12

1,631

4

92

3

16

2

0.31

0.08

0.054

0.056

99.9

28.05

0.06

              

Average

 

28.38

0.47

ACS

75-01

0.2020

0.0003

11,629

11

3,238

6

184

6

18

5

3.14

0.14

0.108

0.023

92.0

1.176

0.006

75-03

0.2020

0.0003

4,741

5

1,419

4

76

3

14

2

0.27

0.09

0.047

0.047

98.5

1.169

0.008

75-04

0.2020

0.0003

11,887

4

3,330

5

166

3

45

3

2.44

0.10

0.111

0.066

94.0

1.195

0.004

75-05

0.2020

0.0003

11,502

6

3,112

4

172

4

5

3

3.72

0.11

0.103

0.007

90.3

1.189

0.004

75-06

0.2020

0.0003

12,491

5

3,668

3

198

3

15

2

0.90

0.10

0.122

0.020

97.9

1.186

0.003

75-08

0.2020

0.0003

16,880

19

3,793

7

215

6

17

6

14.91

0.30

0.126

0.022

73.5

1.170

0.010

79-01

0.1910

0.0002

13,275

12

3,623

6

245

6

22

5

1.48

0.14

0.120

0.026

96.8

1.194

0.005

79-02

0.1910

0.0002

13,908

12

3,866

6

209

6

25

5

0.39

0.14

0.128

0.027

99.3

1.202

0.004

79-03

0.1910

0.0002

25,059

10

6,147

5

382

3

41

4

11.51

0.15

0.204

0.033

86.3

1.187

0.003

79-04

0.1910

0.0002

17,716

6

4,857

5

275

3

37

4

1.96

0.10

0.161

0.037

96.8

1.189

0.003

79-05

0.1910

0.0002

14,645

5

3,944

4

241

3

17

3

2.89

0.11

0.131

0.021

94.2

1.177

0.003

79-06

0.1910

0.0002

7,592

5

2,118

5

130

5

14

4

0.63

0.10

0.070

0.032

97.6

1.178

0.006

79-07

0.1910

0.0002

11,883

6

3,176

5

175

3

19

3

2.72

0.09

0.105

0.029

93.2

1.175

0.004

79-08

0.1910

0.0002

10,674

5

2,921

3

207

3

18

2

1.68

0.10

0.097

0.031

95.4

1.173

0.004

79-09

0.1910

0.0002

7,120

6

1,973

3

117

4

18

4

0.33

0.10

0.066

0.043

98.8

1.200

0.006

83-01

0.1788

0.0002

21,443

11

5,512

7

415

6

35

4

3.32

0.14

0.183

0.027

95.5

1.172

0.003

83-02

0.1788

0.0002

13,561

12

3,278

6

286

7

23

4

4.76

0.15

0.109

0.030

89.6

1.171

0.005

83-03

0.1788

0.0002

18,842

6

4,812

5

305

4

26

3

1.93

0.11

0.160

0.026

97.0

1.199

0.003

83-04

0.1788

0.0002

25,856

5

6,779

6

406

4

40

3

1.77

0.11

0.225

0.029

98.1

1.180

0.002

83-05

0.1788

0.0002

27,183

9

5,713

7

489

6

24

5

19.39

0.31

0.190

0.021

78.7

1.186

0.005

83-06

0.1788

0.0002

12,743

9

3,243

7

187

5

12

5

2.09

0.27

0.108

0.018

95.1

1.180

0.008

83-07

0.1788

0.0002

12,061

14

3,075

6

174

6

13

5

2.18

0.27

0.102

0.021

94.6

1.172

0.009

83-08

0.1788

0.0002

18,775

18

4,755

7

272

6

21

6

2.12

0.27

0.158

0.022

96.7

1.205

0.006

83-09

0.1788

0.0002

17,799

18

4,521

7

269

6

18

6

1.97

0.27

0.150

0.020

96.8

1.202

0.006

83-10

0.1788

0.0002

6,962

17

1,784

6

111

6

14

6

0.44

0.26

0.059

0.040

98.2

1.210

0.015

              

Average

 

1.185

0.004

Figure 3

Age probability diagrams for the 40 Ar/ 39 Ar age of standard materials. (a) Fisher Canyon Sanidine. (b) Alder Creek Sanidine.

For age dating of young rocks (<1 Ma), the exact derivation of radiogenic 40Ar (40Ar*) is indispensable to get meaningful 40Ar/39Ar ages, so that it needs to minimize the reactor-derived 40Ar from 40K. In this study, the (40Ar/39Ar)K with the 0.25-mm-thick cadmium shielding reaches approximately 0.08, which is higher than the majority of reactor facilities used for 40Ar/39Ar age dating (McDougall and Harrison 1999) owing to high slow/fast neutron flux of >780 in Hanaro (Nam S, personal communication). To reduce the incidence of slow neutrons and, thus, the production rate of 40Ar, thicker cadmium shielding should be highly considered.

Conclusions

40Ar/39Ar dating system with multi-collector mass spectrometer and CO2 laser heating device has been established at KBSI in order to date young and very small amounts of minerals and volcanic rocks. Preliminary experiments on sanidine standards show the reliable age results of total-fusion dating for single grains, i.e., 28.4 ± 0.5 Ma (1σ, MSWD = 138, n = 24) for FCS sanidine, and 1.185 ± 0.004 Ma (1σ, MSWD = 6.05, n = 26) for ACS sanidine as Quaternary age standard, within recommended values (28.294 ± 0.036 Ma and 1.193 ± 0.001 Ma, respectively).

Declarations

Acknowledgements

This study was supported by the KBSI grant (G34200). We thank Dr. Alan Deino in Berkeley Geochronology Center and Mr. Karl Burke in Thermo for their effort to install MassSpec software and ARGUS VI mass spectrometer. We appreciate Mr. Nam Seongsu who organized the neutron irradiation process in HANARO. We are also grateful to reviewers who read the manuscripts and give useful comments.

Authors’ Affiliations

(1)
Division of Earth and Environmental Sciences, Korea Basic Science Institute

References

  1. Cheong CS, Kim N, Kim J, Yi K, Jeong YJ, Park CS, Li HK, Cho M (2014) Petrogenesis of Late Permian sodic metagranitoids in southeastern Korea: SHRIMP zircon geochronology and elemental and Nd–Hf isotope geochemistry. J Asian Earth Sci. doi:10.1016/j/jaes.2014.06.005
  2. Faure G, Mensing TM (2005) Isotopes: principles and applications, 3rd edn. John Wiley & Sons, HobokenGoogle Scholar
  3. Mark DF, Barford D, Sturat FM, Imlach J (2009) The ARGUS multicollector noble gas mass spectrometer: performance for 40Ar/39Ar geochronology. Geochem Geophys Geosys. doi:10.1029/2009GC002643
  4. McDougall I (2014) Perspectives on 40Ar/39Ar dating. In: Jourdan F, Mark DF, Verati C (eds) Advances in 40Ar/39Ar dating: from archeology to planetary science, vol 378. Geological Society Special Publication, London, pp 9–20Google Scholar
  5. McDougall I, Harrison TM (1999) Geochronology and thermochronology by the 40Ar/39Ar method, 2nd edn. Oxford Univ. Press, New YorkGoogle Scholar
  6. Nomade S, Renne PR, Vogel N, Deino AL, Sharp WD, Becker TA, Jaouni AR, Mundil R (2005) Alder Creek sanidine (ACs-2): a Quaternary 40Ar/39Ar dating standard tied to the Cobb Mountain geomagnetic event. Chem Geol 218:315–338View ArticleGoogle Scholar
  7. Renne PR, Mundil R, Balco G, Min K, Ludwig KR (2010) Joint determination of 40K decay constants and 40Ar*–40K for the Fish Canyon sanidine standard, and improved accuracy for 40Ar/39Ar geochronology. Geochim Cosmochim Acta 74:5349–5367View ArticleGoogle Scholar
  8. Wilde SA, Valley JW, Peck WH, Graham CM (2001) Evidence from detrital zircons for the existence of continental crust and oceans on the Earth 4.4 Gyr ago. Nature 409:175–178View ArticleGoogle Scholar

Copyright

© Kim and Jeon; licensee Springer. 2015

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.