V INTERNATIONAL ACI/CANMET CONFERENCE ON HIGH PERFORMANCE CONCRETE STRUCTURES AND MATERIALS

PARAMETERS FOR SELF-COMPACTING

CONCRETE MORTAR PHASE

Miguel Nepomuceno
Miguel Nepomuceno is a Civil Engineer and Professor of the Architecture and Civil Engineering Department of the University of Beira Interior in Portugal.

Luiz Oliveira
Luiz Oliveira is a RILEM member and Professor of the Architecture and Civil Engineering Department of the University of Beira Interior in Portugal.

Index
1- Abstract 2 Experimental program 3- Experimental results and discussion 4- Conclusions

1- Abstract
This paper reports an experimental study on the mortar phase for self-compacting concrete.

A series of mortars were produced with similar flow properties, measured by spread and v-funnel tests, adequate to produce selfcompacting concrete.
The water content and the modified carboxylic superplasticizer dosage were determined experimentally for each mortar.
Different percentages of cement replacement materials were used in binary blends, each one combining one of the two types of cement with one of the three mineral additions selected: limestone powder, granite filler and fly ash.

1- Abstract
Each of the binary blends of powders was combined in five different proportions in volume with the fine aggregate (Vp/Vs). Mortars were tested for compressive strength at 28 days age and this value was related to the water/cement ratio, the percentage of replacement materials, and Vp/Vs parameter. The analysis revealed the possibility of establishing adequate mortar parameters to obtain simultaneously the self-compactability and the required compressive strength of self-compacting concrete.

Index
1- Abstract 2 Experimental program 3- Experimental results and discussion 4- Conclusions

2- Experimental Program
Materials used / Powder Materials: Normal portland cement with a specific gravity of 3.14 Calcareous portland cement with a specific gravity of 3.04 Limestone powder with a specific gravity of 2.72 Granite filler with a specific gravity of 2.65 Fly ash with specific gravity of 2.38. Materials used / Admixtures: Modified polycarboxylic based superplasticizer with a density of 1.05.

2- Experimental Program
Materials used / Fine aggregates: The fine aggregate grading curve results from the combination of two sands (40% of Sand 01 and 60% of Sand 05 in absolute volume). Sand 01 was produced from crushed granite with very fine particles and a specific gravity of 2.59 and a fineness modulus of 1.49.

Sand 05 was river sand with a specific gravity of 2.61 and a fineness modulus of 2.71.

2- Experimental Program
Materials used / Fine aggregate grading curve:
Grading curve of fine aggregates
100 90 80 70 60 50 40 30 20 10 0

Material that pass trougth sieve [%] .

Grading curve of combined fine aggregates

Sand 01

Sand 05

0,59

1,19

2,38

4,76

9,52

12,7

19,1

25,4

0,074

0,149

0,297

Aperture of sieve mesh [mm]

38,1

2- Experimental Program
Mix proportions of the powder materials(*)
Table 1 Volume fractions in binary blends of powders
Materials Mixtures
Ref.
Binary mixtures

Cement
CEM I 42.5R (C1) C1 = 3140 kg/m3 ------------fc1= 1.00 fc1= 0.70 fc1= 0.70 fc1= 0.70 fc1= 0.60 fc1= 0.60 fc1= 0.60 fc1= 0.50 fc1= 0.40 CEM II/B L32.5N (C2) C1 = 3040 kg/m3 fc2= 1.00 fc2= 0.80 fc2= 0.80 fc2= 0.80 fc2= 0.60 fc2= 0.50 ------------------Fly ash (CV) CV = 2380 kg/m3 ------fcv= 0.20 ----------fcv= 0.30 ----fcv= 0.40 -----

Addition
Limestone powder ( FC) FC = 2720 kg/m3 --ffc= 0.20 ----ffc= 0.40 ffc= 0.50 --ffc= 0.30 ----ffc= 0.40 ----ffc= 0.50 ffc= 0.60 Granite filler (FG) FG = 2650 kg/m3 ----ffg= 0.20 ----------ffg= 0.30 ----ffg= 0.40 -------

Ref.

1 2 3 4 14 15 5 6 7 8 9 10 11 12 13

100C2 80C2+20FC 80C2+20FG 80C2+20CV 60C2+40FC 50C2+50FC 100C1 70C1+30FC 70C1+30FG 70C1+30CV 60C1+40FC 60C1+40FG 60C1+40CV 50C1+50FC 40C1+60FC

(*) Proportions in terms of the absolute volume

Binary mixtures

2- Experimental Program
Studies in mortars / Parameters for mortar mix design: Proportions of powder materials. Vp/Vs (ratio, in absolute volume, powder materials / fine aggregates). Vw/Vp (ratio, in absolute volume, water content / powder materials). Sp/p% (ratio, in percentage, mass of the superplasticizer / mass of the powder materials). The volume of voids and the contribution of powders from fine aggregates were not considered.

2- Experimental Program
Studies in mortars / Evaluation of fresh properties of mortars: Spread test (Figure 1a) The result is expressed in terms of relative flow area (Gm):
Dm Gm D 1 0
Dm - mean value of the two perpendicular diameters D0 - initial diameter of the base of the cone
Spread test
70 mm

100 mm

Figure 1a Spread test

59 mm

2- Experimental Program
Studies in mortars / Evaluation of fresh properties of mortars: V-funnel test (Figure 1b) The result is expressed in terms of the relative flow velocity (Rm).
29 270 mm

Rm

10 t
30

t - time of flow in the v-funnel test, in seconds.

V-funnel test

Figure 1b V-funnel test

60

240 mm

2- Experimental Program
Studies in mortars / Mixing sequence: Batches of 1.6 liters were produced
Superplasticizer Powders + fine aggregates + water
Mix (normal velocity) Rest (to clean blade) Mix (normal velocity) Rest

Discharge

Mix (normal velocity)

10

11

Time (minutes)

Figure 3 Mixing sequence used to produce mortars

2- Experimental Program
Studies in mortars / Achieve the required fresh properties: Mortars were produced combining each mixture of powder materials shown in Table 1, with different values of the Vp/Vs. The Vp/Vs varied from 0.60 to 0.80. On average, for each combination of powder materials and Vp/Vs, three mortars were produced, varying the Vw/Vp and Sp/p% until required fresh properties were obtained. The experimental procedure is shown schematically on Figs 4 and 5.

2- Experimental Program
Studies in mortars / To achieve the required fresh properties:
Mortar fresh properties

Rm

For Sp/p% constant, the increase of Vw/Vp produces a linear translation of (Gm, Rm) values.
Fig. 4 Mortar fresh properties when Vw/Vp increases with Sp/p% constant

2,00
Increase of Sp/p%

1,80 1,60 1,40 1,20

Increase of Vw/Vp

1,00 0,80
Increase of Vw/Vp with constant value of Sp/p%

0,60 0,40 0,20 0,00 0,00

Sp/p% = constant Vw/Vp = constant Target assumed in this research work

1,00

2,00

3,00

4,00

5,00 Gm

6,00

7,00

8,00

9,00

10,00

2- Experimental Program
Studies in mortars / To achieve the required fresh properties:
Mortar fresh properties

Rm

Figure 5 shows the variation of the (Gm, Rm) values when the Vw/Vp remains constant and Sp/p% increases.
Fig. 5 Mortar fresh properties when Sp/p% increases with Vw/Vp constant

2,00
Increase of Sp/p%

1,80 1,60 1,40 1,20

Increase of Vw/Vp

1,00 0,80
Increase of Sp/p% with constant value of Vw/Vp

0,60 0,40 0,20 0,00 0,00

Sp/p% = constant Vw/Vp = constant Target assumed in this research work

1,00

2,00

3,00

4,00

5,00 Gm

6,00

7,00

8,00

9,00

10,00

2- Experimental Program
Studies in mortars / Achieve the required fresh properties: Admissible range for Gm and Rm is shown in Fig. 6.
Mortar fresh properties

Gm between 5.3 and 5.9

(Dm between 251 and 263 mm)

1,80 1,70 1,60 1,50

Value proposed by Okamura research group [1,2,3 and 4]

Rm

Rm between 1.14 and 1.30 s-1

(t between 7.69 e 8.77 s)

1,40 1,30 1,20 1,10 1,00

Overlapping of different research works in Japan [11]

Fig. 6 Admissible interval of variation for fresh properties of mortars

0,90 0,80 3,00

Target assumed in this research work

4,00

5,00 Gm

6,00

7,00

8,00

2- Experimental Program
Studies in mortars / To achieve the required fresh properties: Subsequent studies have revealed that properties defined for mortars are adequate to produce concretes with values of: Gc between 8 and 11.25 (flow diameter on concrete Dm between 600 and 700 mm)

Rc between 0.5 and 1.0 s-1, (v-funnel time on concrete t between 10 and 20 s).

2- Experimental Program
Studies in mortars / Mortar compressive strength:

For the mortars that satisfied the required fresh properties 4 cubic specimens of 50 mm side were produced and cured in water at 202 C.

Samples were tested for compressive strength at 28 days, perpendicular to the cast direction and using a loading rate of 1.5 kN/s.

Index
1- Abstract 2 Experimental program 3- Experimental results and discussion 4- Conclusions

3- Experimental Results and Discussion

Analysis of mixture proportions / Water content:
Mortars with CEM II/B-L32.5N
275,0 270,0 265,0 260,0 255,0

Fig. 7 Water content versus Vp/Vs for mortars with CEM II/B-L32,5 N
A1 # 100C2 A2 # 80C2+20FC
275,0

A3 A1

Mortars with CEM I 42.5R

A5 # 100C1

W [litres/m 3]

A3 # 80C2+20FG
250,0 245,0

270,0

A2 A4
A4 # 80C2+20CV

A6 # 70C1+30FC
265,0

A10 A5 A7
A8 # 70C1+30CV A7 # 70C1+30FG

240,0 235,0 230,0 225,0 220,0 0,55 0,60 0,65 0,70 0,75 0,80 0,85

260,0 255,0 250,0 245,0 240,0

A14 # 60C2+40FC

A14 A15
A15 # 50C2+50FC

W [litres/m3]

A9 # 60C1+40FC

A6
A10 # 60C1+40FG

A9 A8 A12 A13 A11

225,0 0,55 0,60 0,65 0,70 0,75 0,80 0,85

A11 # 60C1+40CV

Vp/Vs

235,0 230,0

A12 # 50C1+50FC

Fig. 8 Water content versus Vp/Vs for mortars with CEM I 42,5 R

A13 # 40C1+60FC

Vp/Vs

3- Experimental Results and Discussion

Analysis of mixture proportions / Water content: Mortars in which powder proportions include only cement and those in which this cement is partially replaced by granite filler, have the higher water requirements, and clearly detached from the other mortars in which cement was partially replaced by limestone powder or fly ash; The dosage of water tends to decrease with the increase of percentage of cement replacement by the addition, when this addition is limestone powder or fly ash; For the mortars that incorporate granite filler, the results seems to indicate that the dosage of water didnt vary with cement replacement percentage and, in all cases, it was almost equivalent to the dosage of water used in the reference mix with only cement.

3- Experimental Results and Discussion

Analysis of mixture proportions / Superplasticizer content:
Mortars with CEM II/B-L32.5N
25,00

A1 # 100C2

Fig. 9 Superplasticyzer content versus Vp/Vs for mortars with CEM II/B-L32,5 N
Mortars with CEM I 42.5R
35,00

A1
20,00

A3
Sp [litres/m 3]
15,00

A2 A4

A2 # 80C2+20FC

A3 # 80C2+20FG

A5 # 100C1
30,00

A14
10,00

A6 # 70C1+30FC
A4 # 80C2+20CV

A5
25,00

A15
A14 # 60C2+40FC

A7 # 70C1+30FG

A8
Sp [litres/m 3]
20,00

A7 A6 A10 A11 A9

A8 # 70C1+30CV

5,00

A9 # 60C1+40FC A10 # 60C1+40FG

A15 # 50C2+50FC
0,00 0,55 0,60 0,65 0,70 0,75 0,80 0,85

15,00

10,00

Vp/Vs
5,00

A12 A13

A11 # 60C1+40CV

A12 # 50C1+50FC

Fig. 10 Superplasticyzer content versus Vp/Vs for mortars with CEM I 42,5 R

A13 # 40C1+60FC
0,00 0,55 0,60 0,65 0,70 0,75 0,80 0,85

Vp/Vs

3- Experimental Results and Discussion

Analysis of mixture proportions / Superplasticizer content: The higher amounts of superplasticizer always occurs for reference mortars which include only cement. When comparing the mortars with the same cement and the same percentage of cement replacement, the mean value of superplasticizer dosage is similar for all the three additions used.

In mortars that incorporate the same types of cement and addition, the amount of superplasticizer decreases with the increase of percentage of cement replacement by the addition.

3- Experimental Results and Discussion

Analysis of mixture proportions / Mix design parameters:
Mortar parameters
0,85

A1 # 100C2 A2 # 80C2+20FC A3 # 80C2+20FG

A1 2 A1 1 A7

0,80

A4 # 80C2+20CV A5 # 100C1 A6 # 70C1+30FC

A1 0 A5

0,75

A1 3

Vp/Vs

A1 5

A7 # 70C1+30FG A8 # 70C1+30CV A9 # 60C1+40FC A10 # 60C1+40FG A11 # 60C1+40CV A12 # 50C1+50FC A13 # 40C1+60FC A14 # 60C2+40FC

0,70
A1 4 A8 A9 A3 A2 A1 1

0,65
A4 A6

0,60

0,55 0,66 0,68 0,70 0,72 0,74 0,76 0,78 0,80 0,82 0,84 0,86 0,88 0,90 0,92 0,94 0,96 0,98 1,00

A15 # 50C2+50FC

Vw/Vp

Fig. 11- Relationship between Vp/Vs and Vw/Vp for all the mortars produced

3- Experimental Results and Discussion

Analysis of mixture proportions / Mix design parameters: For mortars with the same powder proportions, the increase of Vp/Vs leads to lower volumetric ratio between the water content and the dosage of powder materials (Vw/Vp). For mortars with the same value of Vp/Vs and the same types of cement and addition, the ratio Vw/Vp decreases with the increase of the percentage of cement replacement by the addition.

3- Experimental Results and Discussion

Analysis of mixture proportions / Mix design parameters:
Mortar parameters
0,98 0,96 0,94 0,92 0,90 0,88 0,86
A4 A7 A1 0 A3 A1 A5

A1 # 100C2 A2 # 80C2+20FC A3 # 80C2+20FG A4 # 80C2+20CV A5 # 100C1 A6 # 70C1+30FC A7 # 70C1+30FG

A6

Vw/Vp

0,84 0,82 0,80 0,78 0,76 0,74 0,72 0,70 0,68 0,66 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 2,20 2,40 2,60 2,80 3,00 3,20 3,40 3,60 3,80 4,00
A1 3 A1 5 A9 A1 1 A2 A1 2 A1 4 A8

A8 # 70C1+30CV A9 # 60C1+40FC A10 # 60C1+40FG A11 # 60C1+40CV A12 # 50C1+50FC A13 # 40C1+60FC A14 # 60C2+40FC A15 # 50C2+50FC

Sp/p%

Fig. 12- Relationship between Vw/Vp and the Sp/p% for all the mortars produced

3- Experimental Results and Discussion

Analysis of mixture proportions / Mix design parameters: for mortars with the same binary mixture of powder, the dosage of the superplasticizer, expressed in terms of Sp/p%, didnt change significantly with variation of Vw/Vp. In mortars with the same types of cement and addition, the dosage of the superplasticizer, expressed in terms of Sp/p%, tends to decrease when the percentage of cement replacement by the addition increases.

3- Experimental Results and Discussion

Analysis of mixture proportions / Mortar compressive strength:
Mortar parameters A1 # 100C2
100,0 95,0 90,0 85,0 80,0 75,0 70,0
A1 1 A7 A1 A1 0 A1 2 A3 A1 3 A1 4 A2 A9 A4 A8 A6 A5

A2 # 80C2+20FC A3 # 80C2+20FG A4 # 80C2+20CV A5 # 100C1 A6 # 70C1+30FC A7 # 70C1+30FG A8 # 70C1+30CV A9 # 60C1+40FC A10 # 60C1+40FG A11 # 60C1+40CV A12 # 50C1+50FC A13 # 40C1+60FC A14 # 60C2+40FC A15 # 50C2+50FC
0,60 0,65 0,70 0,75 0,80 0,85

fm,28

65,0 60,0 55,0 50,0 45,0 40,0 35,0 30,0 25,0 20,0 0,55

A1 5

Vp/Vs

Fig. 13- Relationship between the mortar compressive and the Vp/Vs

3- Experimental Results and Discussion

Analysis of mixture proportions / Mortar compressive strength: For mortars with the same combination of powder materials, the compressive strength varied about 10 MPa, when Vp/Vs varied from 0.60 to 0.80. This confirms the initial hypothesis that, besides the powder proportions, the Vp/Vs parameter contributes to the control of mortar compressive strength. The results shown that this relationship clearly exists for each one of the mortars that incorporate the same type of powder association.

3- Experimental Results and Discussion

Mortar parameters

Analysis of mixture proportions / Mortar compressive strength: As expected, the mortar compressive strength decreases with the increase of W/C ratio.
Figure 14 indicates a general relationship for each type of the cement used.
fm,28

100,0 95,0 90,0 85,0 80,0 75,0 70,0 65,0 60,0 55,0 50,0 45,0 40,0 35,0 30,0 25,0
2 1

y = -64.195Ln(x) + 9.9222 R2 = 0.9607 y = -62.392Ln(x) - 14.094 R2 = 0.9089

CEM I 42,5 R

Mixtures with CEM II/B-L32.5N (A1, A2, A3, A4, A14, A15)

Mixtures with CEM I 42.5R (A5, A6, A7, A8, A9, A10, A11, A12, A13)

CEM II/B L32,5N

W/C

Fig. 14- Compressive strength versus W/C ratio, expressed in terms of mass.

20,0 0,22 0,26 0,30 0,34 0,38 0,42 0,46 0,50 0,54 0,58 0,62 0,66

3- Experimental Results and Discussion

Analysis of mixture proportions / Mortar compressive strength: Fig. 15 to Fig. 20 shows the relationship between the percentage of cement replacement by the addition (fAd.), the W/C ratio and Vp/Vs. The results presented from Fig. 15 to Fig. 20 can be used together with the relationship between compressive strength and W/C ratio (Fig. 14) to estimate all the mixture parameters of any mortar that includes the materials analyzed in this research work. It is believed that, even if a different polycarboxylic based superplasticizer is used, the variations produced on water content will be not so large so as to significantly affect the W/C ratio presented in this research work.

3- Experimental Results and Discussion

Analysis of mixture proportions / Mortar compressive strength:
Relationship between fAd. and W/C
0,70

Unit percentage of cement replacement by the addition (fAd.)

0,65 0,60 0,55 0,50 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05

CEM I 42.5R + LIMESTONE POWDER

0.70 Vp/Vs = 0,60

Fig. 15 Relationship between (fAd.), W/C ratio and Vp/Vs for combination of CEM I 42,5 R and Limestone powder
Relationship between fAd. and W/C

0.80 0.60

Vp/Vs = 0,65

Vp/Vs = 0,70 0.75 0.65

Unit percentage of cement replacement by the addition (fAd.)
0,70 0,65 0,60 0,55 0,50 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00 0,240 0,280 0,320 0,360 0,400 0,440 0,480 0,520 0,560 0,600 0,640 0,680

Vp/Vs = 0,75

CEM I 42.5R + GRANITE FILLER

Vp/Vs = 0,60 0.80 0.75 0.70 0.65 0.60 Vp/Vs = 0,70 Vp/Vs = 0,65

Vp/Vs = 0,80

0,00 0,240

0,280

0,320

0,360

0,400

0,440

0,480

0,520

0,560

0,600

0,640

0,680

W/C

Fig. 16 Relationship between (fAd.), W/C ratio and Vp/Vs for combination of CEM I 42,5 R and Granite filler

Vp/Vs = 0,75

Vp/Vs = 0,80

W/C

3- Experimental Results and Discussion

Analysis of mixture proportions / Mortar compressive strength:
Relationship between fAd. and W/C
0,70 0,65 0,60 0,55 0,50 0,45 0,40 0,35 0,30 CEM I 42.5R + FLY ASH

Unit percentage of cement replacement by the addition (fAd.)

Vp/Vs = 0,60 0.75 0.70 0.65 0.60 Vp/Vs = 0,70 Vp/Vs = 0,65

0.80

Fig. 17 Relationship between (fAd.), W/C ratio and Vp/Vs for combination of CEM I 42,5 R and Fly ash
Relationship between fAd. and W/C 0,60

Unit percentage of cement replacement by the addition (fAd.)

0,25 0,20 0,15 0,10 0,05 0,00 0,240 0,280 0,320 0,360 0,400 0,440 0,480 0,520 0,560 0,600 0,640 0,680

0,55 0,50 0,45 0,40 0,35 0,30

Vp/Vs = 0,75

CEM II/BL 32.5N + LIMESTONE POWDER

0.70

Vp/Vs = 0,60

Vp/Vs = 0,80

0.80

Vp/Vs = 0,65

Vp/Vs = 0,70 0.60

W/C

0,25 0,20 0,15 0,10 0,05 0,00 0,240

Fig. 18 Relationship between (fAd.), W/C ratio and Vp/Vs for combination of CEM II/B L32,5N and Limestone powder

0.75 0.65

Vp/Vs = 0,75

Vp/Vs = 0,80

0,280

0,320

0,360

0,400 W/C

0,440

0,480

0,520

0,560

3- Experimental Results and Discussion

Analysis of mixture proportions / Mortar compressive strength:
Relationship between fAd. and W/C
0,60

Unit percentage of cement replacement by the addition (fAd.)

0,55 0,50 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00 0,240

CEM II/BL 32.5N + GRANITE FILLER

Vp/Vs = 0,60

Fig. 19 Relationship between (fAd.), W/C ratio and Vp/Vs for combination of CEM II/B L32,5N and Granite filler
Relationship between fAd. and W/C
0,60

Vp/Vs = 0,65

0.80

0.75

0.70 0.65

Vp/Vs = 0,70
Unit percentage of cement replacement by the addition (fAd.)

0.60 Vp/Vs = 0,75

0,55 0,50 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00 0,240

CEM II/BL 32.5N + FLY ASH

Vp/Vs = 0,60

Vp/Vs = 0,65

Vp/Vs = 0,80

0,280

0,320

0,360

0,400 W/C

0,440

0,480

0,520

0,560

0.80

0.75

0.70 0.65 0.60

Vp/Vs = 0,70

Vp/Vs = 0,75

Fig. 20 Relationship between (fAd.), W/C ratio and Vp/Vs for combination of CEM II/B L32,5N and Fly ash

Vp/Vs = 0,80

0,280

0,320

0,360

0,400 W/C

0,440

0,480

0,520

0,560

Index
1- Abstract 2 Experimental program 3- Experimental results and discussion 4- Conclusions

4- Conclusions
The following conclusions can be drawn based on the results of this experimental investigation for the mortar mixtures and procedures used: Comparing mortars with the same value of Vp/Vs and the same types of cement and addition, the ratio Vw/Vp decreases with the increase of the percentage of cement replacement by the addition. For mortars with the same powder proportions, the dosage of the superplasticizer, expressed in terms of Sp/p%, doesnt change significantly with variation of Vw/Vp. Mixtures with the same Vp/Vs values and the same types of cement and addition have shown that the dosage of the superplasticizer, expressed by Sp/p%, tends to decrease when the percentage of cement replacement by the addition increases.

4- Conclusions
The comparative analysis of mortars doesnt reveal a general relationship between the analyzed parameters Vp/Vs, Vw/Vp and Sp/p%. However, it can be useful to preview trial mixtures behavior in laboratory to obtain the required fresh properties.
Good agreement was observed, for each powder association mortar, between the mortar compressive strength and the Vp/Vs parameter. The mortar compressive strength increases as Vp/Vs ratio increases. As expected the compressive strength decreases with the increase of W/C ratio for mortars with the same type of cement.

For binary mixtures incorporating a certain type of cement and addition, it is possible to find a good correlation between the W/C ratio and the percentage of cement replacement by the addition when the Vp/Vs is constant.

4- Conclusions
It can be concluded that, after having selected the type of powder materials to be used and the required mortar compressive strength, it is possible to estimate the adequate W/C ratio. Moreover, from W/C ratio and Vp/Vs, it is possible to estimate the required percentage of the cement replacement by the addition.