SILICA FUME AND GLASS POWDER Student: Supervisors: Raymond - - PowerPoint PPT Presentation

OPTIMIZING HIGH PERFORMANCE SELF COMPACTING CONCRETE BY ADDING SILICA FUME AND GLASS POWDER

Student: Supervisors: Raymond Alexander Yonathan

• Prof. Dr. Ir. Triwulan, DEA.

3114202003 3114202003

• Dr. Eng. Januarti JE, ST., MT.

1

Contents

1. Background 2. Purpose 3. Define controlled factors, levels, degree of freedom, and orthogonal arrays 4. Materials 5. Mix Design 6. Test Results 7. Taguchi data analysis 8. Discussions 9. Conclusion

2

Background

3

Developing countries Construction technologies Concretes

More cement productions

Green house effect Durable concretes

Background

Silica Fume and Glass Powder as Cement Replacement

• Improved strength and durability of concrete
• Glass powder is suitable for self compacting concrete

High performance self Compacting Concrete

• Efficient, productive, durable, good quality
• Less human errors

Taguchi’s data analysis method

• Efficient experimental activities

4

Purpose

• 1. Optimizing high performance self compacting concrete as the

main purpose

• 2. Analyzing the effects of materials used in high performance

self compacting concrete

5

Define controlled factors, levels, degree of freedom, and orthogonal arrays

Code Controlled Factors Level 1 Level 2 Level 3 (%) (%) (%) A Coarse Aggregate 45 50 55 B Glass Powder 10 15 20 C Silica Fume 20 40 60 D Viscocrete 0,8 1 1,2

Factors Degree of Freedom A 2 B 2 C 2 D 2 Total 8

6

• Coarse aggregate from aggregate’s volume
• Glass powder from cement’s weight
• Silica fume from glass powder’s weight
• Viscocrete from binders weight
• Degree of Freedom = Number of Levels - 1

Define controlled factors, levels, degree of freedom, and orthogonal arrays

Orthogonal Arrays L9(34) Exp Code A B C D R1-10-20 1 1 1 1 R2-15-40 1 2 2 2 R3-20-60 1 3 3 3 R4-10-40 2 1 2 3 R5-15-60 2 2 3 1 R6-20-20 2 3 1 2 R7-10-60 3 1 3 2 R8-15-20 3 2 1 3 R9-20-40 3 3 2 1

7

La(bc) Where: L = Latin Square Design a = Number of rows b = Number of levels c = Number of factors

Materials

1. Ordinary Portland Cement  PT. Varia Usaha Beton 2. Glass Powder  PT Asahimas Flat Glass Tbk (need to be pulverized using ball mill) 3. Silica Fume  PT BASF Indonesia 4. Water  PDAM ITS Laboratory 5. Sika Viscocrete / polycarboxylate ether  PT SIKA Indonesia 6. Plastimen VZ / Polyhydroxy Carbon Salts PT SIKA Indonesia 7. Coarse Aggregate  PT Surya Beton Indonesia 8. Fine Aggregate  PT Surya Beton Indonesia

8

Mix Design

9

40% paste + 60% aggregate

Exp Code Cement (kg/m3) Glass Powder (kg/m3) Silica Fume (kg/m3) Water (Liter/m3) Viscocrete (liter/m3) Plastimen VZ (Liter/m3) Fine A (kg/m3) Coarse A (kg/m3) R1-10-20 612,75 54,47 13,62 170,21 5,11 1,36 897,6 739,8 R2-15-40 569,85 60,34 40,22 167,6 6,29 1,34 897,6 739,8 R3-20-60 527,27 52,73 79,09 164,77 7,43 1,32 897,6 739,8 R4-10-40 607,73 40,52 27,01 168,81 7,61 1,35 816 822 R5-15-60 570,06 40,24 60,36 167,67 5,04 1,34 816 822 R6-20-20 532,79 106,56 26,64 166,5 6,25 1,33 816 822 R7-10-60 608,53 27,05 40,57 169,04 6,35 1,35 734,4 904,2 R8-15-20 569,63 80,42 20,1 167,54 7,55 1,34 734,4 904,2 R9-20-40 532,5 79,88 53,25 166,41 5 1,33 734,4 904,2

• Yellow Color showed highest value

Test Results

Codes Compressive Strength on 28 Days Slump V-Funnel L-Box Total Porosity (MPa) (cm) (s) (%) R1 56,34 79,5 8,17 0,83 17,93 R2 53,2 72,5 12,3 0,79 15,51 R3 40,98 69,5 7,1 0,83 20,89 R4 42,93 71 11,8 0,76 16,11 R5 45,99 68,5 14 0,73 24,57 R6 42,76 80,5 8,3 0,82 18,5 R7 32,16 73,5 12,94 0,78 29,02 R8 22,35 79 37,37 0,8 21,95 R9 31,82 71 8,81 0,79 22,71

10

0,00 10,00 20,00 30,00 40,00 50,00 60,00 0,00 5,00 10,00 15,00 20,00 25,00 30,00 35,00 R1 R2 R3 R4 R5 R6 R7 R8 R9

Varian Compressive Strength (MPa) Porosity (%)

Porositas Total Porositas Terbuka Porositas Tertutup Kuat Tekan

Badly Bubbled ±5cm Bleeding

• Around 1 cm thick paste showed at the surface of specimen  need to be cut and converse before tested

Taguchi data analysis (85% confidence level)

Test Coarse Aggregate Glass Powder Silica Fume Viscocrete Estimated optimum Condition Slump flow (cm) 55% ** 10% 20% * 1% * 78,778 ± 2,040 cm L-Box 55% * 15% * 40% * 1% ** 0,760 ± 0,030 V-Funnel (s) 45% 20% 40% ** 1%

• Compressive

Strength (MPa) 45% * 10% 40% ** 0,8% * 54,037 ± 5,496 MPa Porosity (%) 45% * 15% ** 40% * 1,2% * 14,274 ± 0,677 % Note: (*) Significant Factor (**) Pooled Factor

11

Single response

Test Coarse Aggregate Glass Powder Silica Fume Viscocrete Estimated optimum Condition Slump flow (cm) 45%* 15%** 40%* 1%* 73,056 ± 2,040 cm L-Box 0,807 ± 0,032 V-Funnel (s)

• Compressive

Strength (MPa) 53,828 ± 5,314 MPa Porosity (%) 15,636 ± 0,677 % Note: (*) Significant Factor (**) Pooled Factor

Multi response

• V-Funnel’s optimum condition is unclear  no significant factors  optimum condition will have a big error

Discussion

• A. Slump Flow

Optimum: Coarse aggregate 55%, Glass Powder 10%, Silica Fume 20%, Viscocrete 1% (single-response). Coarse aggregate 45%, Glass Powder 15%, Silica Fume 40%, Viscocrete 1% (multi-response). Estimated: 78,778 ± 2,040 cm (single-response) 73,056 ± 2,040 cm (multi-response) Significant factors: the greatest effect is Silica Fume, followed by viscocrete. (single-response) the greatest effect is Silica Fume, followed by coarse aggregate, and then viscocrete. (multi-response)

• At R6 and R8 bleeding occurred  highest amount glass powder  cannot reduce bleeding.
• Liu (2010), Limantono (2015), Ekaputri (2015)  Glass Powder absorb less amount of water than silica

fume.

12

Discussion

• B. V-Funnel

Optimum : - Estimated : - Significant factors : -

• Unclear result  wrong assumption of factors or levels.

13

Discussion

• C. L-Box

Optimum: Coarse aggregate 55%, Glass Powder 15%, Silica Fume 40%, Viscocrete 1% (single-response). Coarse aggregate 45%, Glass Powder 15%, Silica Fume 40%, Viscocrete 1% (multi-response). Estimated: 0,760 ± 0,030 (single-response) 0,807 ± 0,032 (multi-response) Significant factors: the greatest effect is Coarse aggregate, followed by silica fume, and then glass powder. (single-response) the greatest effect is Silica Fume, followed by coarse aggregate, and then viscocrete. (multi-response)

• Gravel has important role (Okamura, 2003; Brouwers, 2005)  by lowering coarse aggregate, could

increase segregration resistance as shown by L-box results.

14

Discussion

• D. Compressive Strength

Optimum: Coarse aggregate 45%, Glass Powder 10%, Silica Fume 40%, Viscocrete 0,8% (single-response). Coarse aggregate 45%, Glass Powder 15%, Silica Fume 40%, Viscocrete 1% (multi-response). Estimated: 54,037 ± 5,496 MPa (single-response) 53,828 ± 5,314 MPa (multi-response) Significant factors: the greatest effect is Coarse aggregate, followed by viscocrete. (single-response) the greatest effect is Silica Fume, followed by coarse aggregate, and then viscocrete. (multi- response)

• Lower coarse aggregate  higher fine aggregate  small porous  high strength (Rashid, 2009; Logan,

2009)

• Same recommendation composition (Limantono,2015; Ekaputri, 2015)  10-15% Glass Powder, 40-60%

silica fume.

15

Discussion

• E. Porosity

Optimum: Coarse aggregate 45%, Glass Powder 15%, Silica Fume 40%, Viscocrete 1,2% (single-response). Coarse aggregate 45%, Glass Powder 15%, Silica Fume 40%, Viscocrete 1% (multi-response). Estimated: 14,274 ± 0,677 % (single-response) 15,636 ± 0,677 % (multi-response) Significant factors: the greatest effect is Silica fume, followed by coarse aggregate, and then viscocrete. (single-response) the greatest effect is Silica Fume, followed by coarse aggregate, and then viscocrete. (multi-response)

• It was found that (Wille, 2011) using high range water reducer and microsilica could decreased

porosity.

16

Conclusion

1. Taguchi’s method provide more result with less experimental trial. Using bigger orthogonal array, better result could be achieved. 2. For V-funnel test, another trial need to be done  misassumption on the factors and levels cause unclear result. 3. Glass Powder absorb small amount of water. 4. Gravel has great effect on segregration. 5. It is recommended to use 10-15% of glass powder, 40% of silica fume, 0,8% of viscocrete  high compressive strength. 6. Using higher range water reducer and microsilica can lower the concrete porosity. 7. Another trial should be done regarding the combination of viscocrete + plastimen vz + glass powder  causing bubble at certain amount (assumption).

17

18

THANK YOU