Newsprint Research

• This is a summary of research conducted at the Smithsonian Center for Materials Research and Education (SCMRE) during the summer of 2003. • This work was conducted by two interns, Evan Quasney and Kathy Hufford, under the supervision of David Erhardt and Charles Tumosa. • Drs. Erhardt and Tumosa may be contacted at SCMRE via these methods:

Charles Tumosa W. David Erhardt [email protected] [email protected] (301) 238-3700 x 118 (301) 238-3700 x 116

Permanence and Degradation: Newsprint Over the Last 100 Years

Kathy Hufford Massachusetts Institute of Technology

Evan Quasney University of Michigan

Charles Tumosa, Ph.D. W. David Erhardt, Ph. D. Introduction ƒ Printed material on paper dominates written communication

ƒ Paper records have finite life … how finite?

ƒ Mechanical concepts addressed and discussed

ƒ Chemical concepts linked to mechanical properties

ƒ Scope of research newsprint- specific Wood-based Newspapers Tested 06/01/2003 Washington Post 11/17/2001 Washington Post 11/27/2000 Washington Post 08/03/1999 Washington Post 12/02/1998 Washington Post 11/02/1997 Washington Post 07/10/1997 Washington Post 01/26/1997 Washington Post 05/07/1995 Washington Post 11/14/1993 Washington Post 02/08/1983 Washington Post 10/20/1983 Washington Post 07/07/1985 Washington Post 09/11/1985 Washington Post 12/14/1988 Washington Post 12/15/1988 Washington Post 10/25/1999 Washington Post 07/01/1975 Vineland Times Journal 12/11/1960 New York Times 09/13/1950 Christian Science Monitor 12/02/1934 Topeka Daily Capital 04/11/1915 Detroit Free Press 10/07/1905 Detroit Free Press 01/03/1890 The World 05/07/1875 New York Semi-Weekly Times Definitions

Strain = ∆ length / length Stress = Force / Area - extensibility of paper - ‘strength’ of the paper - change in length - randomized based on ε = ∆ L / L sizing or technology σ = F / A

Stress-Strain Curve Tensile (Young’s) Modulus - Graph of stress versus - Numeric value of the strain flexibility / stiffness of the - Basis for finding plastic paper and elastic regions E = σ / ε Definitions

Isotropic: Breaking Strain: - Specimen behaves the - Percent elongation at which same in all directions a specimen fails

Orthotropic: Breaking Stress: - Specimen behaves - Pressure at the breaking differently in mutually strain; tensile strength of perpendicular directions specimen

Region Deformation: Elastic – Flexible Modulation of Specimen Plastic – Permanent Irreparable Damage to Specimen General Stress-Strain Curve Mancellinus Antonius, circa 1500, Vertical Dir 25

20 Slope = E

Plastic Deformation 15

σ = 10.25

Stress (MPa) 10 Elastic Deformation 5 ε = 0.004

0 0 0.005 0.01 0.015 0.02 0.025 Strain (mm/mm) Method

ƒ Tests performed on screw-driven tensile tester in environmental chambers ƒ Incremental length change standard: 30-seconds, 1/200 (0.005) inches ƒ Tests performed between 42 – 52% RH and 22.5 – 24.2 deg. C ƒ 4 distinctly different sub-variations of each specimen examined Machine v. Cross Direction

Directional Comparison of Century Magazine 10 circa Feb. 1925 8

6

4

Stress (MPa) 2 0 0 0.003 0.006 0.009 0.012 0.015 Strain (mm/mm) Machine Dir Cross Dir Initial Application Tensile Strength of The Washington Post 25 June 1, 2003, Average Value, Cross & Machine Dir.

20

15

10 Stress (MPa) 5

0 0 0.002 0.004 0.006 0.008 0.01 Strain (mm/mm) Inked Uninked Initial Application

Individual Axial Comparison of Inked v. Uninked Paper The Washington Post, June 1, 2003 5 25

20 4

15 3

10 2

5 Uninked 1 Uninked Inked Inked 0 0 0 0.002 0.004 0.006 0.008 0.01 0 0.005 0.01 0.015 0.02 0.025 0.03 Machine Direction Cross Direction

ƒ Orthotropic behavior present in specimens ƒ Very little difference between inked and uninked strains 20 Year Study in Lab Environment 0.03 0.025 0.02 0.015 0.01 0.005 Strain (mm/mm) 0 0 5 10 15 20 25 Age (Years) Mach. Inked Mach. Uninked Cross Uninked Cross Inked ƒ No great change, but a slight decreasing trend is exhibited ƒ Degradation small enough that overall damage is minimal ƒ Specimens could last 100-150 years 20 Year Study in Lab Environment 0.03 0.025 0.02 0.015 0.01 0.005 Strain (mm/mm) 0 0 5 10 15 20 25 Age (Years) Mach. Inked Mach. Uninked Cross Uninked Cross Inked ƒ No great change, but a slight decreasing trend is exhibited ƒ Degradation small enough that overall damage is minimal ƒ Specimens could last 100-150 years 20 Year Study in Lab Environment

35 28 21 14

Stress (MPa) Stress 7 0 0 5 10 15 20 25 Age (Years) Mach. Inked Mach. Uninked Cross Uninked Cross Inked

ƒ No distinguishable change in strength of specimens ƒ Specimens will last a long time under Standard Laboratory Conditions 20 Year Study in Lab Environment

35 28 21 14

Stress (MPa) Stress 7 0 0 5 10 15 20 25 Age (Years) Mach. Inked Mach. Uninked Cross Uninked Cross Inked

ƒ No distinguishable change in strength of specimens ƒ Specimens will last a long time under Standard Laboratory Conditions Inked v. Uninked Paper

T.S. Comparison of The New York Times 8 Dec. 11, 1960, Cross dir. 7 6 5 4 3

Stress (MPa) Stress 2 Uninked 1 Inked 0 0 0.002 0.004 0.006 0.008 0.01 0.012 Strain (mm/mm) 120 Year Variable Test Strain v. Time Vert. Uninked Vert. Inked Horz. Uninked Horz. Inked 0.03 0.025 0.02 0.015 0.01

Strain (mm/mm) Strain 0.005 0 020406080100120 Age (Years) • Loss of elastic region occurs when breaking strains drops below 0.005 • Acute fragility not present until samples are 80 years old • Total specimen disintegration only occurs after ALL breaking strains are less than 0.003 Scientific Significance

ƒ Newsprint can last 100 years if given nominal attention

ƒ Archival facilities can provide up to an additional 50 – 100 years of viable storage

ƒ Flexibility and elasticity data will help build an accurate model to predict degradation Chemical Composition of Wood

Wood

Lignin Extractives Softwood: 25% Carbohydrates 2-8% Hardwood: 21%

Cellulose Hemicellulose 45% Softwood: 25% Hardwood: 35% Glucose Xylan, Xylose, Arabinose, etc. Cellulose: Glucose Glucose Glucose Glucose Glucose

Glucose Trimer:

Glucose Glucose Glucose

Glucose Dimer:

Glucose Glucose

CH -OH Glucose: 2 O OH

HO OH OH Xylan:

Xylose Xylose Xylose Xylose Xylose Arabinose

Arabinose

Xylose: Arabinose: CH -OH 2 CH2-OH O HO O OH HO OH OH OH Hydrolysis of Cellulose

Glucose Glucose Glucose Glucose Glucose

H2O H2O H2O

Glucose Monomer

Glucose Glucose Dimer

Glucose Glucose Glucose Trimer Method

ƒSamples prepared, extracted in water, filtered, and evaporated under vacuum ƒDerivatized with STOX (commercial reagent containing internal standard), HMDS, and trifluoroacetic acid ƒSupernatant analyzed by gas chromatography ƒSugars identified by comparison of retention times against internal standards Standard

Xylose

Xylose Dimer Glucose Glucose Dimer Arabinose Xylose Trimer Glucose Trimer Approximate Retention Times of Sugar Peaks

Standard: 16.70

Glucose: 12.57

Xylose: 10.43

Arabinose: 10.20

Glucose Dimer: 19.45

Glucose Trimer: 24.55

Xylose Dimer: 17.45

Xylose Trimer: 22.55 Sugar Content in Wood-based Newspapers 4500

4000

3500

3000

2500 Glucose 2000 Xylose

Micrograms/g 1500

1000

500

0 1860 1880 1900 1920 1940 1960 1980 2000 2020 Year ƒGlucose and xylose levels highest in newspapers from the Industrial Revolution ƒOf interest are the spikes around World War I and the relative stability of the past 20 years History of Papermaking Technology

ƒThe Industrial Revolution (1875-1950) brought commercial use of wood-pulp paper and mass production processes

ƒCommercial use of the acid sulfite process (1880s) and Kraft process with bleaching (1930s) caused a transition from mechanical to chemical processing

ƒGlucose and xylose levels peak in the World War I era

ƒAfter World War I, new technology and processing techniques were invented Glucose Fraction in Wood-based Newspapers

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3

Glucose/(Glucose+Xylose) 0.2 0.1 0 1860 1880 1900 1920 1940 1960 1980 2000 Year ƒPre-1980 glucose ratio shows a different mechanism than post-1980 ƒKinetics study? ƒChanges in data from 1980 to present are particularly interesting History of Papermaking Technology

•The USA Today Effect: Launching of the USA Today in 1982 and the use of color in newspapers

•The environmental movement and recycling of newspapers

•Advanced machinery requires thinner material

•Multitude of processes for newspaper manufacturers to choose from: refiner, chemical, thermo, chemothermo, isothermo, etc.

•Further research Sugar Polymers Detroit Free Press, 1905 4500

4000

3500

3000

2500 Glucose Polymers 2000 Xylose Polymers

1500

Sugar, Micrograms/gram 1000

500

0 01234 Polymer ƒXylose is more hydrolytic and more likely to hydrolyze to monomer than is glucose Arabinose and Xylose in Wood-based Newspapers

500 450 400 350 300 250 250 200 200 150 150 100 50 100 0 Arabinose, Micrograms/gram Arabinose, 50 0100 0 0 1000 2000 3000 4000 5000 6000 Xylose, Micrograms/gram ƒInitial, sharp increase in arabinose, and then nearly constant value ƒArabinose molecules must be located at the end of the molecule or on branches, and are hydrolyzed first Total Measured Sugar Content vs. Breaking Strain 14000

12000

10000

8000

6000

Micrograms/gram 4000

2000

0 0 0.005 0.01 0.015 0.02 0.025 0.03 Breaking Strain

ƒSimilar trends appear across other sugar data ƒIllustrates the presence of a surface phenomenon Glucose Polymers vs. Breaking Strain Monomer Dimer

4000 2500

3200 2000

2400 1500

1600 1000

800 500 Dimer, Micrograms/gram Glucose, Micrograms/gram Glucose, 0 0 0 0.005 0.01 0.015 0.02 0.025 0.03 0 0.005 0.01 0.015 0.02 0.025 0.03 Breaking Strain Breaking Strain Trimer

1600

1200

800

400 Trimer, Micrograms/gram

0 0 0.005 0.01 0.015 0.02 0.025 0.03 Breaking Strain Hydrolysis Mechanism Conclusions

ƒ Issue of dueling factors of technology and time in the process of degradation

ƒ Xylan hydrolysis vs. cellulose hydrolysis

ƒ Order of degradation, illustration of mechanistic details of hydrolysis

ƒ Surface phenomenon

ƒ Ideas for further research in kinetics and the history of papermaking technology Summary

ƒ Mechanical and physical properties of specimens directly related to hydrolysis of cellulose and hemicellulose

ƒ Hydrolysis of cellulose and hemicellulose affects breaking strain and plasticity of specimen

ƒ Remaining life span of specimen can be estimated via sugar content analysis and/or mechanical testing